Part II · Capital as Regime Artifact

II.A — Braudel's Three Floors

9 min read · 1,702 words

Capitalism is not one single system, but two. One is transparent—the market economy. The other is opaque—the upper level where the great predators roam.

— Fernand Braudel, The Wheels of Commerce (1982)

Part I established that energy and conversion efficiency do substantial explanatory work in the history of growth. But energy alone does not determine economic structure; what matters is who controls the energy, and how that control is institutionalized. This is where capital enters the story—not as a thing, but as a relationship whose form is determined by what constrains the economy.

In 1602, the Dutch East India Company received a charter granting it a monopoly on trade between the Netherlands and everything east of the Cape of Good Hope. The charter gave the Company authority to build forts, maintain armies, negotiate treaties, and coin money—powers normally reserved for sovereign states. Its investors were not buying shares in a trading venture; they were buying shares in a quasi-governmental entity that could set prices, exclude competitors, and deploy violence to protect its margins. The Company’s returns came not from producing goods more efficiently than rivals but from occupying a position that no rival could contest. This was not market competition; this was something else entirely—capital as position rather than capital as production.

The word “capital” carries more weight than its users usually intend. In ordinary speech it means something like “accumulated wealth available for investment,” and in economic models it means something like “a factor of production that is not labor and not land.” Both usages treat capital as a thing—a stock of resources that can be measured, priced, and deployed. But the history of capitalism suggests that capital is less a thing than a relationship, and that the relationship changes its character depending on what constrains the economy at any given moment.

Fernand Braudel, the French historian associated with the Annales school, spent most of his career trying to describe that relationship in its full historical thickness.1 His magnum opus, Civilization and Capitalism, 15th–18th Century, is an attempt to write the history of capitalism not as a sequence of ideas or institutions but as a set of layered structures, each operating on a different timescale and according to different rules. Braudel’s model divides economic life into three tiers.

At the bottom is what he calls “material life”—the routines of daily existence: food, clothing, shelter, tools, the rhythms of agriculture and craft production. This layer changes slowly, over generations and centuries, and the constraints here are physical: the yield of the land, the availability of fuel, the limits of human and animal muscle.

Above material life sits the “market economy”—the layer of exchange, prices, and visible trade. This is where goods move from producer to consumer, where prices are set by supply and demand, where competition disciplines behavior and allocates resources. The market economy is more dynamic than material life; it responds to shocks, adjusts to scarcity, and transmits information through prices.

But Braudel insists that capitalism proper is neither material life nor the market economy. Capitalism, in his usage, is the third floor: the zone of large-scale, long-distance trade, finance, and concentrated market power, where the rules of ordinary exchange are suspended or bent by those with sufficient resources to escape them. This is the realm of the great merchant houses, the chartered trading companies, the bankers who lend to princes, the operators who corner markets, and the rentiers who live off accumulated claims. Here, prices are not set by competition but by position; information is asymmetric; and the line between commerce and politics dissolves. The Dutch East India Company was a third-floor entity. So were the Medici bank, the British East India Company, and the railroad trusts of the Gilded Age. Braudel’s point is not conspiratorial but structural: the third floor operates by different logic than the market below it.

The distinction matters because it reframes what we mean by “capital.” In the market economy, capital is productive: it funds workshops, buys inventory, improves land, and generally makes things that people want. In Braudel’s third floor, capital is something else—a claim on the surplus generated by the lower floors, a position that allows its holder to extract value without necessarily producing anything. The merchant who buys cheap in one port and sells dear in another is not adding to the world’s stock of goods; he is arbitraging a discontinuity in information or transport that allows him to capture a margin. The banker who lends to a war-making prince is not funding production; he is betting on the prince’s ability to tax his subjects or seize resources elsewhere. The operator who corners the grain market is not feeding anyone; he is positioning himself to charge whatever the traffic will bear.

This is not a moral judgment, or at least not primarily one. Braudel is describing a structural feature of capitalism as it has actually existed. His point is that the third floor operates by different logic than the market economy, and that confusing the two leads to analytical error. The market economy is governed by competition, which tends to erode profits and discipline inefficiency. The third floor is governed by the search for positions that escape competition—monopolies, privileged access, political protection, asymmetric information—and it tends to concentrate rather than disperse wealth. The two layers coexist, but they pull in different directions.

Giovanni Arrighi, an Italian economic sociologist who built on Braudel’s framework, traced this dynamic through four centuries of capitalist development.2 His book The Long Twentieth Century (1994) argues that capitalism has moved through a series of “systemic cycles of accumulation,” each centered on a different hegemonic power: the Italian city-states in the fifteenth and sixteenth centuries, the Dutch Republic in the seventeenth, Britain in the nineteenth, and the United States in the twentieth. Each cycle begins with a phase of material expansion, in which capital is invested in production and trade, and ends with a phase of financial expansion, in which capital withdraws from production and seeks returns through lending, speculation, and the manipulation of money itself. The shift from material to financial expansion is not a sign of health; it is a sign that the productive opportunities within the existing regime have been exhausted, and that capital is searching for new outlets that the current structure cannot provide.

Arrighi’s framework suggests that what counts as “capital” is not fixed but regime-dependent. In a phase of material expansion, capital means factories, ships, railroads, and the organizational capacity to coordinate large-scale production. In a phase of financial expansion, capital means liquidity, creditworthiness, and the ability to move money across borders faster than regulators can follow. The underlying capacity—command over resources—remains, but its form changes as the constraints of the era change.

Karl Polanyi, writing a generation before Braudel, made a related point from a different angle.3 His book The Great Transformation (1944) argues that the self-regulating market—the idea that society should be organized around markets for land, labor, and money—is not a natural state of affairs but a historical anomaly, imposed by deliberate policy in nineteenth-century England and subsequently extended, with varying success, to the rest of the world. Polanyi calls land, labor, and money “fictitious commodities” because they are not produced for sale the way ordinary goods are. Land is the physical environment; labor is human activity; money is a token of purchasing power. Treating them as commodities requires stripping them of their social and ecological context and pretending that they can be bought and sold like bolts of cloth.

The relevance to Braudel’s framework is this: the boundary between the market economy and capitalism proper is not fixed but politically constructed. What counts as a legitimate market transaction and what counts as rent-seeking depends on the institutional framework within which exchange takes place. A medieval guild restricting entry to a trade looks like a monopoly to a free-market economist, but it also provided quality control, training, and social insurance to its members. A modern patent looks like a legitimate reward for innovation, but it also grants a temporary monopoly that allows its holder to charge prices far above marginal cost. The line between productive capital and extractive capital is drawn by law, custom, and power, not by some natural distinction inherent in the assets themselves. Computation may prove to be the next fictitious commodity—not produced for sale in any ordinary sense, but increasingly treated as if it were a fungible factor of production with a price that captures its value.

This has implications for how we think about the present. If capital is the permission to relax the binding constraint, then the question for any era is: what is the binding constraint, and who holds the permission to relax it? In an agrarian economy, the constraint is land and the labor to work it; capital means control over territory and the people who till it. In an industrial economy, the constraint is energy and the machinery to convert it; capital means control over coal mines, steel mills, railroads, and the organizations that coordinate them. In a financial economy, the constraint is liquidity and the creditworthiness to access it; capital means control over balance sheets, payment systems, and the institutions that create and destroy money.

The argument of this manuscript is that the binding constraint is shifting again. The previous part established that energy and conversion efficiency do substantial explanatory work in the history of growth, and that the Solow residual is in part a thermodynamic residual. The next section will argue that the current transition—the rise of computation as a factor of production—fits the pattern of previous regime shifts, and that the assets that mattered in the last regime may not be the assets that matter in the next one. Braudel’s three floors remain; what changes is who controls the staircase between them—and in the current transition, the staircase is being rebuilt out of silicon, electricity, and the permission to run inference at scale.