An artificial person
In 1819 Chief Justice John Marshall of the United States Supreme Court gave a definition that lawyers still quote. A corporation, he wrote, is "an artificial being, invisible, intangible, and existing only in contemplation of law." It exists because a legal system agrees to treat a group of people and their pooled property as a single unit that can act in its own name.
That single unit is a legal entity separate from its owners. It can own land and machines, sign contracts, borrow money, hire workers, and sue or be sued, all without its owners signing every document themselves.
The core features
Most business corporations share a handful of features. The first is legal personality, just described. The second is limited liability: shareholders can lose no more than they invested. If the company collapses owing millions, creditors can take what the company owns, but they cannot take a shareholder's house. The third is transferable shares, which owners can sell without asking permission from the other owners. The fourth is perpetual succession: the company continues when its owners change, whether they sell, retire or die. The fifth is centralized management. In a typical business corporation the shareholders elect a board of directors, and the board hires and oversees the managers who run the business day to day.
The separateness of the company can be startling. In 1897 the House of Lords heard the case of Aron Salomon, a London boot maker who turned his business into a company and held almost all of its shares, with family members holding one share each. When the company failed, creditors argued that Salomon and the company were really the same person, so he should pay its debts. The Lords disagreed: a company is a separate person even when one individual controls it. They make exceptions, a practice called piercing the corporate veil, mainly when the company is a sham used to commit fraud or dodge an existing duty.
Why limited liability matters
Defenders call limited liability one of the great legal inventions. Without it, a retired teacher who owned a few shares in a railway could be ruined by the railway's debts, so cautious savers would avoid companies they could not watch closely. With it, millions of strangers can each put in small sums, spread their savings across many firms, and let professional managers do the work.
Critics point to who carries the risk instead. Banks and suppliers who lend to a company can charge for the danger or demand guarantees. But some creditors never chose to deal with the firm at all: neighbors poisoned by a chemical leak, or passengers injured by a defective product. For these people, limited liability means that victims who never agreed to deal with the firm may be left unpaid when it goes bankrupt.
Who really controls it
In 1932 the lawyer Adolf Berle and the economist Gardiner Means published a study of the largest American companies. They found that shares had become scattered across huge numbers of small investors, none of whom owned enough to steer the business. As a result, they argued, control of big firms had passed to salaried managers. Ownership and control had come apart.
Later economists named the resulting difficulty the agency problem. Shareholders are the principals and managers are their agents, but managers may serve themselves rather than the owners, through lavish perks, empire building or a quiet life. Many familiar features of corporate life are attempts to close this gap: pay in shares, independent directors, audits, the threat of a takeover, and activist investors who buy a stake and demand change.
Many shapes
The giant company with shares traded on a stock exchange is only one kind of corporation. Cities, universities, charities, churches and cooperatives are often incorporated too. What they share is the legal tool, not the goal. A corporation is something people build to get things done together. Its personhood is a legal convenience, which is worth remembering whenever a company and a human being are described as if they had the same standing.
Worked example
Maya and Kofi run a bicycle repair shop as a partnership. Each is personally responsible for all of the shop's debts, so when a customer slips on a wet floor and sues, both of their savings are at risk. If they incorporate, the shop becomes its own legal person. The lawsuit is now against the company, and their personal losses are capped at what they put in, unless a court finds the company was a sham.
Two years later they need money to open a second shop. Their aunt buys a quarter of the shares. She can sell them later to someone else, and if Kofi moves abroad the company carries on. But the aunt now faces an agency problem in miniature. Maya and Kofi decide how to spend the money, and they might pay themselves generous salaries rather than dividends. To protect herself, she asks for a seat on the board and an annual audit. In one small shop, the whole logic of the corporation is visible: separation, limited risk, pooled capital, and the constant question of who is watching the people in charge.