The receipt is a small document on which to rest a constitution. It records what the customer paid, and the saving can be real: a larger factory, a better route or a more reliable system may put something within reach that was previously too expensive. Yet the receipt does not record every condition of the bargain. It does not say whether a supplier can reach another buyer, whether a worker can leave for another employer, or whether a business must please the proprietor of the road by which its customers arrive. These are economic questions before they become political ones. A power that governs access can alter the terms on which other people make their living even while one visible price falls. The law's difficulty is to distinguish the achievement that deserves room from the restraint that deserves an answer, without making success itself the offense.
The American statutes begin with different wrongs rather than a general prohibition of large enterprises. The Sherman Act of 1890 addresses restraints of trade and monopolization; the first concerns combinations, while the second can reach the conduct of a single firm. Section 7 of the Clayton Act, enacted in 1914 and broadened in 1950, addresses acquisitions whose effect may substantially lessen competition or tend to create a monopoly. Its language looks ahead. A merger need not have completed the destruction of competition before that provision can reach it. These distinctions matter because a cartel, the exclusion of a rival and the purchase of a competitor are different ways of acquiring or exercising power. Calling all three bigness discards the mechanism the law must examine.
A successful monopolist is not, merely by being successful, a statutory monopolizer. In United States v. Grinnell, the Supreme Court distinguished possession of monopoly power accompanied by its willful acquisition or maintenance from growth attributable to a superior product, business skill or historical accident. The distinction places a burden on the explanation. A high share may tell the investigator where to look; it does not by itself tell the investigator what happened. Nor does an attractive product settle how the firm treats a rival that threatens its position. The same business can have created something valuable and later used a separate restraint to defend the returns. The task is to identify the restraint and its consequences, not to choose between calling the whole enterprise admirable and calling it guilty.
The older concern about concentrated power nevertheless reached beyond the price of the next purchase. In Brown Shoe, decided in 1962, the Court read the 1950 amendment to the Clayton Act against Congress's concern about rising concentration, local control and the survival of competitive market structures. It also recognized that mergers could strengthen competition and that the law protected competition rather than the fortunes of a particular competitor. Both propositions belonged to the same account. A small firm could be inefficient; a large firm could be the very rival needed to check another large firm. The historical quarrel was therefore already about the kind of economic order the law should preserve and the evidence by which it should recognize a threat. It was never made simple by arranging companies in order of size.
Robert Bork sought to make the choice of purpose much narrower. His 1966 argument treated consumer welfare as the governing value of the Sherman Act and understood that value through economic efficiency: restrictions on output counted against it, while productive improvements counted in its favor. This was more than a preference for a cheap sticker. It was an effort to give judges a common standard and to prevent the same conduct from being condemned or excused according to an unannounced preference for one kind of business over another. The attraction is substantial. A court that can invoke any social concern without explaining how it bears on the statutory wrong can make the law difficult to anticipate. Bork's account tried to turn an expansive political vocabulary into a discipline of economic consequences.
The common standard still has to decide which consequences are being added. Suppose a hundred units continue to be sold, with their quality unchanged. Each formerly cost eight to produce and sold for ten. A new arrangement reduces production cost to six but raises the selling price to eleven. Resources used fall by two hundred. Buyers pay a hundred more. The seller's profit rises by three hundred: two hundred from lower cost and a hundred transferred from buyers. With quantity fixed and no other effects, the monetary gains counted across both sides rise by two hundred while buyers become worse off. A price of nine instead would give buyers a hundred of the saving. Neither calculation establishes what a court should do. They show why a gain in total surplus, a gain for purchasers and a reduction in production cost are three propositions requiring their own evidence.
The people hidden by a narrow account are not always consumers who paid too much. A firm can buy labor, raw materials or access to land as well as sell a finished product. If it obtains better terms by making production more efficient, there may be a real saving to share. If it obtains them by removing an alternative employer or buyer, the lower payment may reflect a weakened position on the other side. A cheap final product does not tell us which occurred. Nor can an accounting transfer to the firm establish a resource saving. The inquiry has to follow the affected market far enough to discover what alternatives remain and whether the changed bargain alters output, investment or the conditions under which people work.
Lina Khan's 2017 critique brought that problem of structure to the platform. Her argument was that measures centered on price and output could miss how a platform acquired control over access and used its position across connected lines of business. A service might be useful and inexpensive while its proprietor also governed the route by which other businesses reached their customers. The immediate advantage would be real; the concern would be what the arrangement enabled or foreclosed. This does not make a period of losses proof of predation or integration proof of exclusion. It changes the questions put to them. Who can enter? On what terms can a dependent seller reach the public? What conduct prevents a rival from becoming an effective alternative? The answers require a mechanism and evidence, not a suspicion attached to the word platform.
The published federal merger framework already reaches well beyond a single retail-price test. The 2023 Merger Guidelines discuss competition in product quality and innovation, as well as competition for workers and other suppliers. Their treatment of buyers asks whether a merger can worsen wages, benefits or other terms, and rejects the idea that harm to competition among employers is answered merely by benefits in a downstream market. In February 2025 the Justice Department expressly retained those guidelines, and its public guidance still listed them in October 2026. They are an enforcement statement, not a replacement for statute or a binding command to a court. Their breadth nevertheless makes the present quarrel more exacting than a choice between caring about price and caring about anything else. It concerns the harms that count, the inferences the evidence can support, and the institutional limits of the remedy.
The wider political concern begins where control of a market becomes control over the conditions of other people's independence. A business that determines access to livelihood may command obedience without issuing anything resembling an order. That gives the fear of concentrated power its force. It also leaves a problem of legal design. If a remedy is justified by a threatened competitive process, it must answer that threat. If the desired rule instead limits economic power because of its influence on public life, the rule needs a public account of that purpose and of the authority to pursue it. Moving between the purposes without saying so makes either one harder to judge. Economic measurement cannot relieve a political community of choosing what it wants its institutions to protect.
The discipline of proof cuts in both directions. It can prevent a useful integration from being broken merely because it is conspicuous. It can also prevent a familiar low price from being treated as a certificate that every surrounding practice is harmless. A remedy that restores access must leave an access that can actually be used; a remedy that creates a rival must leave a rival capable of competing. Otherwise the decree changes a diagram while dependence survives. The same practical inquiry applies when a merger promises efficiencies: the saving must arise from an identifiable change, and its connection to the proposed combination must be explained. A vocabulary of benefits is as capable of concealing power as a vocabulary of power is of concealing benefits.
The receipt can measure a saving without measuring the power behind it.
The law's quarrel is consequently with both the terms of exchange and the arrangements that make those terms possible, in forms that the governing law can recognize and institutions can administer. Price remains an indispensable witness. It cannot testify to every way in which an alternative is removed. The next question is what happens when the check comes from those on the other side of the bargain: when workers organize, buyers combine or an opposing institution acquires the strength to refuse. That is another way to restrain power, and it brings its own question of restraint. The argument remains open.