Here is a question that haunted economists for centuries: Why are diamonds expensive and water cheap?
On its face, the answer should be obvious -- and obviously the reverse of what we observe. Water keeps you alive. Diamonds sit on fingers and catch the light. If value tracks usefulness, water should cost a fortune and diamonds should sell for pennies. Yet walk into any store and the prices tell the opposite story. A one-carat diamond ring might run you five thousand dollars. A glass of water at an American restaurant is free, and half the time nobody drinks it.
This puzzle -- the diamonds-water paradox -- sat at the center of economic thinking for generations, a stone in the shoe of anyone trying to build a coherent theory of value. And its resolution, when it finally arrived, launched what may be the single most important revolution in the history of economics. [C1]
I want to tell you that story. But I also want to tell you a stranger one that runs alongside it -- a story about how one company spent more than a century exploiting the very logic that resolves the paradox, and how that exploitation is now unraveling before our eyes.
The first known person to crack the diamonds-water puzzle was not an economist at all. It was Galileo.
Writing in 1632, in the midst of his great defense of Copernican astronomy, Galileo paused to make a point about the nature of value. It was, in context, a side argument -- part of a dispute about the relative nobility of heavenly and earthly things. But the insight was devastating:
"People ought to remember that if there were as great a scarcity of soil as of jewels or precious metals, there would not be a prince who would not spend a bushel of diamonds and rubies and a cartload of gold just to have enough earth to plant a jasmine in a little pot, or to sow an orange seed and watch it sprout, grow, and produce its handsome leaves, its fragrant flowers, and fine fruit."
And then, even more pointedly:
"It is scarcity and plenty that make the vulgar take things to be precious or worthless; they call a diamond very beautiful because it is like pure water, and then would not exchange one for ten barrels of water."
For Galileo, this was about cosmology. He probably did not know he was stating a foundational principle of economics. But he had put his finger on the essential mechanism: value is not about what a thing is in the abstract. It is about how much of the thing you already have. [C2]
A glass of water is cheap not because water is unimportant, but because most of us have plenty of it. We are not choosing between having water and not having water. We are choosing whether to have a bit more water -- water at the margin, as economists would come to say. And the value of water at the margin, when your tap runs freely, just is not that high.
Diamonds, by contrast, are scarce. Most people own few or none. So each additional diamond sits on a part of the value curve where desire is still intense and supply is thin. Hence the high price.
This is the core of marginalism, and I will give you a simple definition to hold in your mind, even as I warn you it is too simple:
Marginalism is the economic doctrine that we can best understand value by considering how many units of a good or service an individual already has, and then asking how much an additional -- a marginal -- unit would be worth.
Do you already have one horse, or three? The value of the next horse depends entirely on the answer. And from this deceptively modest starting point, you can derive the supply and demand curves that underpin virtually all of modern economics. [C3]
That sounds simple. Maybe too simple. And indeed for centuries most economists were unable to stumble upon this fruitful analytical starting point. Or if they glimpsed the insight -- as the Salamancan theologians of seventeenth-century Spain did, as Galileo did -- they did not see its full import, did not grasp how it could be turned into a systematic engine of scientific advance.
Then, in the early 1870s, something remarkable happened.
The 1870s Discovery
How three economists independently arrived at the same revolutionary insight about value.
Three economists, working independently in three different countries, published essentially the same insight in the same narrow window of time. William Stanley Jevons of Britain. Leon Walras of France and Switzerland. Carl Menger of the Austro-Hungarian Empire. All three addressed the central question of value theory -- how are prices determined in a market economy? -- and all three arrived at the same answer: you cannot understand value without first asking, "How much of this thing does a person already have?" [C4]
The year 1871 is usually taken as the starting gun, though the timing was not perfectly synchronized. Jevons had presented a version of his theory as early as 1862; Menger's Grundsatze der Volkswirtschaftslehre appeared in 1871; Walras published his Elements of Pure Economics in installments beginning in 1874. But the clustering was tight enough and the convergence striking enough that historians gave it a name: the Marginal Revolution.
I like the Marginal Revolution so much I even named my blog -- coauthored with Alex Tabarrok -- after it.
The simultaneous discovery was no coincidence, though it was not exactly coordination either. All three men were responding to the same inadequacies in classical economics, the same nagging failure of the labor theory of value to explain why a diamond costs more than a loaf of bread despite bread requiring more total labor to produce. They were also, in different ways, responding to the mathematical and philosophical currents of their era -- the rise of calculus as a tool of analysis, the utilitarian tradition of Bentham and Mill, the growing ambition to make economics as rigorous as physics. [C5]
What they produced, collectively, was a framework powerful enough to reorganize the entire discipline. Marginalism did not merely resolve the diamonds-water paradox. It provided the logical scaffolding for supply and demand analysis, for the theory of consumer choice, for the pricing of labor, capital, and land. It made modern economics possible.
But -- and this is one of the central arguments of this book -- marginalism is more complex and more interesting than most people believe. There are, it turns out, multiple marginalisms. The simple textbook version I just gave you is one. But there is also marginalism as tautology, marginalism as engineering tool, marginalism as social commentary, marginalism as a framework for understanding risk. The world is full of marginalisms, which is perhaps in slight tension with the core insights of marginalism itself. [C6]
I will get to all of that. First, though, I want to return to the diamonds.
De Beers and Manufactured Scarcity
How one company used marginalist logic to manufacture and maintain artificial scarcity for a century.
Because here is where the story gets delicious.
If the diamonds-water paradox teaches us that diamonds are expensive because they are scarce, the obvious next question is: scarce compared to what? Scarce by nature, or scarce by design?
The answer, for most of the twentieth century, was: by design. Ruthlessly, brilliantly, by design.
In 1888, Cecil Rhodes consolidated several competing diamond mining operations in South Africa into a single entity: De Beers Consolidated Mines. The logic was pure marginalism, even if Rhodes would not have used that word. Diamonds had recently been discovered in enormous quantities at Kimberley, and the flood of supply was crashing prices. If diamonds were going to remain valuable -- if the margin was going to stay where the money was -- someone had to control the supply. [C7]
Rhodes, and after him the Oppenheimer family who took control of De Beers in the 1920s, did exactly that. They bought up or brokered agreements with nearly every significant diamond mine on earth. They created the Central Selling Organisation, a London-based entity through which the vast majority of the world's rough diamonds were funneled. When demand was weak, De Beers stockpiled stones in its vaults rather than let prices fall. When new mines were discovered -- in Siberia, in Australia, in Botswana -- De Beers found ways to bring them into the cartel, or at least to manage the flow.
The operation was, in effect, a century-long exercise in marginal manipulation. De Beers understood that the value of a diamond depends on how many diamonds are already circulating. Their entire business model was to ensure that the answer to that question was: not too many. They kept the world on a part of the demand curve where desire was high and supply was carefully rationed.
And then they did something even more audacious. They manipulated the demand curve itself.
In 1938, De Beers hired the advertising agency N.W. Ayer to run a campaign aimed at the American market. The result, a few years later, was one of the most successful slogans in advertising history: "A Diamond Is Forever." The campaign did not merely sell diamonds. It created a social norm -- the idea that a diamond engagement ring was an essential symbol of love and commitment, that a man should spend two months' salary on one, that to propose without a diamond was to propose inadequately. [C8]
Think about what this accomplished in marginalist terms. De Beers was not just constraining supply to keep the margin tight. They were engineering demand so that each consumer felt the absence of a diamond more acutely -- felt that they were at a point on their personal value curve where the next diamond (the first diamond, the engagement diamond) was desperately needed. They were manufacturing scarcity on one side and manufacturing desire on the other, a pincer movement that kept diamond prices elevated for the better part of a century.
Galileo would have appreciated the irony. The very mechanism he identified -- that scarcity and plenty make the vulgar take things to be precious or worthless -- was being deliberately weaponized by a single company to ensure that an abundant mineral remained classified as precious. [C9]
The Unraveling
How technology and competition began to dissolve De Beers\' carefully constructed monopoly.
The De Beers monopoly began to crack in the late twentieth century. New producers in Russia, Canada, and Australia refused to play by the cartel's rules. Antitrust pressure mounted. By 2000, De Beers had abandoned its strategy of controlling global supply and repositioned itself as a luxury brand.
But the real marginalist earthquake came from an unexpected direction: the laboratory.
Lab-grown diamonds -- chemically, physically, and optically identical to mined diamonds -- began entering the market in significant quantities in the 2010s. By the mid-2020s, they are threatening to do to the diamond market what Galileo's thought experiment always implied was possible. If you could receive plenty more diamonds by turning on your kitchen tap, diamonds probably would not sell for much. You cannot turn on your kitchen tap for diamonds yet, but you can grow them in a machine, and the cost of doing so keeps falling. [C10]
The price of lab-grown diamonds has plummeted accordingly. A one-carat lab-grown stone that sold for roughly $4,000 in 2016 could be had for under $500 by 2024. The margin has shifted. Where once each diamond sat on a steep, scarce part of the value curve, lab-grown stones are flooding the market and pushing us toward the flat part where additional units simply are not worth very much.
The mined-diamond industry has responded with exactly the arguments you would predict. Natural diamonds are "real." They carry stories of geological time. They have emotional resonance that a factory product cannot replicate. These arguments are not entirely wrong -- meaning and narrative do affect value, and marginalism has always acknowledged that preferences are complex. But the arguments are also, transparently, an attempt to re-erect the scarcity barrier that technology has breached. De Beers, having spent a century telling us "A Diamond Is Forever," now needs us to believe that not all diamonds count. [C11]
This is a marginalism lesson hiding in plain sight. The entire arc -- from Galileo's thought experiment about scarce soil and plentiful jewels, through De Beers' century of manufactured scarcity, to the lab-grown disruption that is unraveling the illusion -- is a story about where you sit on the curve, and who gets to decide where that is.
Multiple Marginalisms
The idea of marginalism is far more complex and contested than its simple textbook presentation suggests.
Now, I should be honest with you about what this book is really about, because it is not primarily about diamonds.
The purpose of this book is to take the idea of marginalism -- one of the most important ideas in all of economics -- and show you that it is deeper, stranger, and more contested than you probably think. To suggest that what is central in economic reasoning is historically contingent, not as obvious as we might have once thought. To give you new ways of thinking about why intellectual progress can be so slow, then extremely fast, and later so evanescent.
I also want to help you see how many complementary pieces must be in place before a new method moves into its heyday. The three founders of 1871 did not invent marginalism from nothing. Galileo had the insight in 1632. The Salamancan theologians had versions of it. The French engineer Jules Dupuit presented a remarkably complete marginalist framework in 1844, including the first systematic case for marginal cost pricing, which he applied to public bridges. Hermann Heinrich Gossen stated the principles clearly in his 1854 German-language treatise -- and was so pleased with himself that he compared his achievement to that of Copernicus. [C12]
Yet none of these earlier efforts caught fire. The insight kept appearing and disappearing, like a match struck in a room without kindling. What changed in the 1870s was not the insight itself but the surrounding infrastructure -- mathematical tools, philosophical frameworks, institutional incentives, the growing ambition of economics to become a real science. Marginalism needed all of those things to go from a clever observation to a revolution.
As the theory developed, it postulated an equilibrium where the marginal utility you get from spending money in different areas is equalized. In formal terms: the ratio of marginal utility to price should be the same across all goods. If an extra dollar spent on beans gives you more satisfaction than an extra dollar spent on cheese, you should keep buying beans and cutting back on cheese until the returns even out. In equation form:
MU(x)/P(x) = MU(y)/P(y) = MU(z)/P(z)
This is elegant, and it is powerful. From this simple principle you can derive models of consumption, theories of wages and rents and returns to capital. You can build the mathematical machinery of modern economics. Marginalism became, in essence, a full employment project for economists -- it gave them tools, it gave them equations, it gave them things to measure and test and argue about. [C13]
But I want to press on something that gets less attention. Why did it take so long? If the insight is as straightforward as I have just made it sound, why did two centuries pass between Galileo and the Marginal Revolution? Why were the Salamancans ignored? Why did Dupuit's brilliant 1844 work fail to ignite the profession?
Part of the answer is that marginalism is not, in fact, as simple as it first appears. I mentioned that there are multiple marginalisms. Let me briefly flag them, because they will structure much of what follows.
There is intuitive marginalism -- the deployment of marginal reasoning to generate surprising, often counterintuitive insights about human behavior. Why do drivers in China sometimes intentionally kill the pedestrians they hit? Because under certain compensation schemes, the fine for a dead victim is lower than the cost of lifetime care for a surviving one. At the margin, you lower your penalty if the victim dies. That is horrifying, and it is marginalism. [C14]
There is tautological marginalism -- the recognition that any choice, examined deeply enough, can be expressed as an equation of values at the margin. This version is unfalsifiable and therefore less exciting as an empirical claim, but it is extraordinarily useful as a framework for building models and generating testable hypotheses.
There is engineering marginalism -- the application of marginal reasoning to improve policy. Congestion pricing. Carbon taxes. Spectrum auctions. Singapore, where even the fines for public misbehavior include a consumption tax because, well, if you are deriving pleasure from the fined activity, that pleasure ought to be taxed like any other form of consumption.
There is social marginalism -- the use of marginal reasoning to make comparisons of well-being across individuals. An extra thousand dollars means more to a poor person than to a rich one, and from this observation flow arguments about redistribution, welfare policy, and the moral weight of economic inequality.
And there is marginalism as it applies to risk -- the insight that diminishing marginal utility implies risk aversion, which gave birth to modern theories of insurance and financial asset pricing.
Each of these marginalisms has its own history, its own champions, its own blind spots. They do not contradict each other exactly, but they have different emphases and sometimes pull in different directions. The simple textbook presentation -- value depends on how much you already have -- papers over a remarkable amount of internal complexity.
Did the simplest presentations of marginalism actually prepare us for the reality that marginalism is a many-splendored thing? I do not think so. And that is part of why, even today, marginalism is both more powerful and more fragile than most people realize.
I hope to explain why ideas we love, and work with fruitfully, can disappear and leave center stage. You and I may be crying, but scientific life will go on. I have already put down my handkerchief.
The diamonds taught us that scarcity determines value. De Beers taught us that scarcity can be manufactured. Lab-grown diamonds are teaching us that manufactured scarcity can be undone. And the history of marginalism itself -- its slow gestation, its sudden flowering, its quiet contemporary decline -- teaches us that even the most fundamental ideas in economics are more historically contingent, more fragile, and more fascinating than we ever suspected.
Let us begin.