1. The Planning Structure
A common transfer pricing plan assigns a fixed role to each member of a multinational group. A contract manufacturer, a limited risk distributor, or a retailer is labeled a “routine” entity and is promised a low operating profit margin inside the interquartile range (IQR) of a set of independent comparables, for example 1% to 4% of net sales. A year-end adjustment enforces the promise. The residual operating profit, positive or negative, flows to an “entrepreneurial” entity, usually the contractual owner of the intangibles, often resident in a low-tax jurisdiction.
Write Z(i) for the operating profit of routine entity i, S(i) for its net sales, and \mu for the target margin chosen inside the IQR. The plan imposes:
\quad (1)\qquad Z(i) = \mu\, S(i), \qquad 0.01 \le \mu \le 0.04
and the entrepreneurial entity E collects the group residual profits:
\quad (2)\qquad Z(E) = Z(G) - \sum_i \mu\, S(i)
where Z(G) is consolidated operating profit and the sum runs over the routine entities. Equation (2) is an accounting identity. It has no behavioral content; it allocates whatever the group reports after the routine entities receive their stipulated return.
This note argues that the structure is unrealistic on two grounds. First, in the competitive sector the theory of competition from Marx through Schumpeter to Cochrane denies that any firm can hold a guaranteed margin; competition compresses margins and punishes laggards. In the oligopoly sector the markup is stable, as the operating markup of 71 large U.S. corporations from 1950 to 2024 demonstrates, but that stability is a property of the degree of monopoly of the group, not of a routine function.
The oligopoly sector has two stages: innovating corporations that renew their degree of monopoly through R&D and advertising, and mature corporations that hold it through concentration. The comparables that set the IQR belong to the mature stage: oligopolies drawn from Compustat, whose low margins reflect low expenditure on innovation and sales promotion. Second, the contribution analysis of such groups shows that research and development (R&D) and advertising expenses, the outlays that create trade and marketing intangibles, are booked in the accounts of the “routine” entities. The entities running the treadmill are the entities the plan calls routine.
2. Schumpeter: Creative Destruction
Schumpeter attributes economic development to the carrying out of “new combinations”: new goods, new methods of production, new markets, new sources of supply, and new forms of industrial organization (Schumpeter 1934, ch. II). Entrepreneurial profit is the surplus that accrues to the innovator while price still reflects the old cost. It is temporary. Imitators arrive in swarms, the new method becomes customary practice, and the surplus is competed away (Schumpeter 1934, ch. IV).
In Capitalism, Socialism and Democracy Schumpeter replaces the textbook picture of price competition with competition from the new commodity, the new technology, and the new type of organization. He names the process:
“the perennial gale of creative destruction” (Schumpeter 1950, p. 84)
“This process of Creative Destruction is the essential fact about capitalism.” (Schumpeter 1950, p. 83)
The threat runs before any rival appears. Incumbents must innovate because the prospect of displacement disciplines conduct in advance:
“It disciplines before it attacks.” (Schumpeter 1950, p. 85)
Two features of Schumpeter’s analysis matter for transfer pricing. First, no position is secure; the operating profit margin of every firm is exposed to displacement. Second, Schumpeter defines the entrepreneur by function, the carrying out of new combinations, not by the bearing of risk. In The Theory of Economic Development the risk of failure falls on the owner of the means of production or the provider of credit; the entrepreneur bears it only when he is also the owner or the lender (Schumpeter 1934, ch. II). The entrepreneur of transfer pricing planning, defined by contractual risk assumption and legal title to intangibles, is not Schumpeter’s entrepreneur.
3. Marx and Cochrane: The Treadmill
Marx described the mechanism before Schumpeter. The capitalist who first adopts an improved method captures an extra surplus value—as long as the old market price is constant. The gain does not last. Marx writes that competition forces competitors to adopt the new method; once the higher productivity becomes the social norm, the extra surplus value disappears (Marx 1976, ch. 12, pp. 433–436). The individual capitalist must innovate again to restore a squeezed profit margin.
The term “treadmill” belongs to the agricultural economist Willard Cochrane (1958, ch. 5). Cochrane’s family farmer adopts a cost-reducing technology and earns a profit while the market price reflects the old cost. As adoption spreads, aggregate output rises, the price falls, and the profit vanishes. Farmers who adopt late suffer losses and exit; the survivors absorb their land. The gains of technical change pass to consumers by lower prices. The farmer must run to remain in place. Cochrane restated the argument in his history of American agriculture (Cochrane 1979), and Levins and Cochrane (1996) revisited it four decades after its first statement.
Cochrane’s model presupposes price-taking firms facing a market price they do not set. That presupposition confines the treadmill to the competitive sector of the economy. The next section draws the boundary.
4. Two Sectors: Competition and Oligopoly
Kalecki (1954, ch. 1; 1971, ch. 5) divides prices into two classes. Demand-determined prices govern raw materials and agricultural products, where supply is limited in the short period and price absorbs the variation in demand. Cost-determined prices govern manufactured goods, where reserve capacity makes supply elastic and the firm sets price as a markup on prime cost. Hicks (1974) calls them flexprice and fixprice systems. Cochrane’s farmers belong to the first class; the large corporation belongs to the second.
In the cost-determined sector the price equation is:
\quad (3)\qquad P(j) = (1 + k)\, C(j)
where C(j) is unit cost of goods sold (COGS), the prime cost of Kalecki, and k is Kalecki’s markup. Kalecki attributes the markup to the degree of monopoly, which depends on industrial concentration, the development of sales promotion through advertising and selling agents, the level of overhead relative to prime cost, and the strength of trade unions (Kalecki 1954, ch. 1).
Silva (2026) estimates the aggregate counterpart of (3) for 71 large U.S. corporations from 1950 to 2024. Write S for revenue (REVT), X for operating expense (XOPR), where X = C + \mathrm{XSGA}, the sum of COGS and selling, general and administrative expense, and Z for operating profit before depreciation (OIBDP). Each corporation satisfies the identity:
\quad (4)\qquad S = X + Z
and the behavioral equation:
\quad (5)\qquad Z = \alpha + \beta S + \varepsilon
Substituting (5) into (4) and solving for S gives the reduced form:
\quad (6)\qquad S = \lambda_0 + \lambda_1 X + u
with \lambda_0 = \alpha/(1-\beta), \lambda_1 = 1/(1-\beta), and u = \varepsilon/(1-\beta).
From (4) and (6), Z = \lambda_0 + (\lambda_1 - 1) X + u, and the markup on operating cost X is \lambda_1 - 1 = \beta/(1-\beta). This markup extends Kalecki’s k from prime cost C to operating cost X; the two coincide when \mathrm{XSGA} = 0.
The slope \beta, and with it \lambda_1 and the markup on X, is fixed across seven decades of technical change, product turnover, and recessions. The treadmill does not compress it. These corporations run to keep their degree of monopoly, not to defend a margin that competition erodes.
Equation (5) also exposes the defect of the IQR. Apply it to a cross-section of comparables i = 1, \ldots, n and divide by S(i): the company-level operating margin is:
\frac{Z(i)}{S(i)} = \beta + \frac{\alpha}{S(i)} + \frac{\varepsilon(i)}{S(i)}.With E(\varepsilon \mid S) = 0, the expected value of the mean margin is \beta + \alpha/H, where H is the harmonic mean of S(i) across the comparables. The quartiles of company-level margins carry the intercept term; they do not measure the slope \beta.
Schumpeter reached a compatible conclusion. In chapter VIII of Capitalism, Socialism and Democracy, he argues that the large-scale enterprise is the engine of technical progress, and that
“perfect competition is not only impossible but inferior” (Schumpeter 1950, p. 106)
The oligopoly sector has two stages. In the innovating stage, creative destruction proceeds through new products and new processes financed out of a stable markup, and continued outlays on R&D and advertising maintain the degree of monopoly that sustains a high markup. In the mature stage, concentration sustains the degree of monopoly, outlays on innovation and sales promotion are small, and the markup is low but no less stable (Section 5). The treadmill is the law of the competitive sector. In either stage of the oligopoly sector the markup persists; its level depends on the stage.
5. The Comparables: Mature Oligopolies
The comparables that populate the IQR are not small firms of the competitive sector. They are oligopolies taken from Compustat: mature corporations that hold a dominant market position through concentration, spend little on advertising, and report low levels of R&D. Their operating margins are low. A dominant position joined to a low margin is a puzzle, and the stage of the product cycle resolves it. These corporations occupy the near-horizontal section of the S-curve; revenue (REVT) grows at low rates, and growth, when it occurs, comes through horizontal acquisition of rivals, not through innovation. The acquisition of Family Dollar by Dollar Tree in 2015 is representative: the acquirer bought market share and store count, not a new product or a new method of production. In Kalecki’s terms, concentration raises the degree of monopoly while the absence of sales promotion holds it down; the markup that results is low and stable. Consolidation, not innovation, defends the position against Schumpeter’s gale.
The profit margins of mature oligopolies remain market outcomes, not stipulations. Across firms and years, the margins disperse: some earn above the third quartile, some below the first, and some report operating losses and become delisted. Their margins have positive variance, and risk is variance. The transfer pricing plan selects a point \mu from the distribution of those outcomes and converts it into a contract with zero variance. No comparable in the Compustat set holds such a guarantee; each bears the variance the plan removes from the tested party. A guaranteed margin is an insurance arrangement, not a market result.
The plan borrows its level from mature oligopolies with low intangible intensity and applies it to an entity whose accounts carry the outlays that create intangibles. The comparables earn a low markup because they have ceased to innovate and to advertise; the tested party books the R&D and advertising that sustain the degree of monopoly of its group. Kalecki names sales promotion among the determinants of the degree of monopoly. A margin taken from firms that spend little on sales promotion cannot measure the return to an entity that spends a material share of its revenue on it. The guarantee draws its stability from the consolidated markup of the group; the routine entities produce that markup, and the plan credits it to the entrepreneurial entity.
The contradiction has two parts. The first is a dilemma on risk. If the routine entity bears no variance, the comparables that bear variance are not comparable in risk. If the comparables are “sufficiently similar” to the routine entity (Treas. Reg. §1.482-1(d)(2)), the routine entity must bear variance, and equation (1) cannot hold year after year. The second is a difference in function: the routine entity books R&D and advertising at rates the mature comparables do not.
6. The Persistence of the Residual
In the competitive sector, innovation profit (nonroutine profit in transfer pricing lingo) decays unless renewed; Schumpeter’s imitators and Cochrane’s late adopters erode it. In the innovating stage of the oligopoly sector the markup persists, as the fixed slope \beta of equation (5) demonstrates, but its persistence depends on continuous outlays: new products from R&D and sustained brand presence from advertising. Where those outlays cease, the group passes to the mature stage, and its markup settles at the low level of the comparables. Either way, a residual that persists in the entrepreneurial entity requires the renewal of the intangibles that generate it.
Transfer pricing rules recognize these expenses. R&D, advertising, and software development expenses carry out the development, enhancement, maintenance, protection, and exploitation (DEMPE) functions, and the OECD Guidelines attribute the return to intangibles to the entities that perform and control those functions, not to the legal owner alone (OECD 2022, ch. VI, Section B). The U.S. regulations treat the same outlays in one of two ways. Where the entity that incurs them is not the legal owner of the intangible, the outlays are a contribution to the value of an intangible owned by another controlled party, a service that requires arm’s length compensation (Treas. Reg. §1.482-4(f)(4)). Where the entity is the owner under the legal owner rule, the outlays develop its own intangibles, and the return to those intangibles is its own (Treas. Reg. §1.482-4(f)(3)). Neither treatment supports a stipulated margin read from comparables that do not incur such outlays.
When those outlays are booked in the routine entities, the persistence of the principal’s residual depends on the routine entities’ spending. Remove their R&D and advertising, and the group passes to the mature stage, where the markup falls to the low level of the comparables, or drifts toward the competitive sector, where the treadmill compresses the margin. The routine entities maintain the markup; the entrepreneurial entity collects the residual. That allocation describes self-dealing contracts among related parties, not economic functions. The U.S. regulations respect contractual terms only where they are consistent with the economic substance of the transaction (Treas. Reg. §1.482-1(d)(3)(ii)(B)), and the OECD Guidelines delineate the actual transaction from the conduct of the parties (OECD 2022, ch. I, Section D.1). An allocation that the accounts of the tested parties contradict fails both tests.
7. Conclusion
The theory of competition from Marx (1976 [1867]), Schumpeter (1934 [1911], 1950 [1942]), and Cochrane (1958) holds that, in the competitive sector, margins are transient and exposed. In the oligopoly sector Kalecki’s markup is stable: high where outlays on R&D and advertising renew the degree of monopoly, low where concentration alone sustains it. The 71 large U.S. corporations of Silva (2026) show a fixed slope from 1950 to 2024. The typical “comparables” that set the IQR are mature oligopolies on the flat section of the S-curve: low margins, low R&D and advertising, growth by horizontal acquisition. Their margins still disperse across firms and years.
A guaranteed operating margin read from their IQR is incoherent on two counts: it removes from the tested party the variance the comparables bear, and it imports the low markup of firms that have ceased to build intangibles into an entity that books the R&D and advertising of its group. Where the routine entities book those expenses, they perform the Schumpeterian function and produce the Kaleckian markup that the plan reserves for the entrepreneurial entity. The contribution analysis and the economic substance of the transaction, not the intercompany contract, should determine where the residual is earned.
References
Cochrane, W. W. (1958). Farm Prices: Myth and Reality. Minneapolis: University of Minnesota Press.
Chapter 5, “The Agricultural Treadmill,” introduces the treadmill: early adopters gain, diffusion lowers price, laggards exit, gains pass to consumers.
Cochrane, W. W. (1979). The Development of American Agriculture: A Historical Analysis. Minneapolis: University of Minnesota Press.
Restates the treadmill as the organizing mechanism of American agricultural history.
Hicks, J. R. (1974). The Crisis in Keynesian Economics. Oxford: Basil Blackwell.
Distinguishes fixprice markets, where producers set prices and output adjusts to demand, from flexprice markets, where prices adjust to clear supply and demand.
Kalecki, M. (1954). Theory of Economic Dynamics: An Essay on Cyclical and Long-Run Changes in Capitalist Economy. London: George Allen & Unwin.
Chapter 1, “Costs and Prices”: cost-determined versus demand-determined prices; the markup on prime cost and the factors determining the degree of monopoly, including sales promotion through advertising.
Kalecki, M. (1971). Selected Essays on the Dynamics of the Capitalist Economy, 1933–1970. Cambridge: Cambridge University Press.
Chapter 5, “Costs and Prices”: the final statement of the pricing equation and the degree of monopoly.
Levins, R. A., and W. W. Cochrane (1996). “The Treadmill Revisited.” Land Economics, 72 (4): 550–553.
Reassesses the treadmill four decades after its first statement.
Marx, K. (1976 [1867]). Capital: A Critique of Political Economy, Vol. I. Trans. B. Fowkes. Harmondsworth: Penguin.
Chapter 12, “The Concept of Relative Surplus-Value” (pp. 433–436): extra surplus value of the first adopter, its generalization (diffusion) through new entrants competition, and its eventual disappearance.
OECD (2022). OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations. Paris: OECD Publishing.
Chapter I, Section D.1: delineation of the actual transaction from the contractual terms and the conduct of the parties. Chapter VI, Section B: the development, enhancement, maintenance, protection, and exploitation (DEMPE) functions and the return to intangibles.
Schumpeter, J. A. (1934 [1911]). The Theory of Economic Development. Trans. R. Opie. Cambridge, MA: Harvard University Press.
Chapter II defines the entrepreneur by the carrying out of new combinations and separates that function from risk-bearing; Chapter IV treats entrepreneurial profit as a temporary surplus eroded by imitation.
Schumpeter, J. A. (1950 [1942]). Capitalism, Socialism and Democracy, 3rd ed. New York: Harper & Brothers.
Chapter VII, “The Process of Creative Destruction” (pp. 81–86): competition from the new commodity and the new technology; the threat that disciplines before it attacks. Chapter VIII, “Monopolistic Practices” (p. 106): the large-scale enterprise as the engine of progress; perfect competition as impossible and inferior.
Silva, E. (2026). U.S. Corporate Profits, 1950–2024. Cham: Springer.
Identity REVT = XOPR + OIBDP and behavioral equation \mathrm{OIBDP} = \alpha + \beta\,\mathrm{REVT} + \varepsilon for 71 large U.S. corporations from 1950 to 2024; reduced-form parameters \lambda; the slope is fixed across the period.
U.S. Treasury Regulations, Section 1.482-1(d)(2) and (d)(3)(ii)(B). Standard of comparability; identifying contractual terms.
An uncontrolled transaction must be sufficiently similar to provide a reliable measure of an arm’s length result; contractual terms are respected only if consistent with the economic substance of the transaction.
U.S. Treasury Regulations, Section 1.482-4(f)(3)–(4) (2009). Ownership of intangible property; contribution to the value of intangible property owned by another.
Legal owner rule and the arm’s length consideration due for contributions to an intangible owned by another controlled party.