What Is Limited Liability? The Legal Fiction That Built Modern Business

Before it existed, business was conducted through partnerships in which each partner was personally liable for everything the business owed, without limit, so a bad year could take a partner's house. That is workable when partners know each other and can watch what the others do, and it does not scale, because no rational person invests a modest sum in a venture run by strangers if the downside is unlimited. Early limited arrangements existed in Roman and medieval commercial law and in the Italian commenda, and the modern form emerged through chartered companies granted privileges by the state, including the Dutch and English East India companies. Britain generalised it in stages, with the Joint Stock Companies Act of 1844 allowing incorporation by registration rather than by special act, and the Limited Liability Act of 1856 attaching limited liability as a standard feature; France and the American states followed in the same period. The legal principle was settled decisively in England by the House of Lords in Salomon against Salomon in 1897, which held that a company is a separate person even when one man owns almost all of it, a decision that horrified many at the time.

The consequences run well beyond the liability rule itself:

• •The company owns its assets, not the shareholders, so a shareholder cannot take a piece of equipment because they own a tenth of the business
• •The company contracts in its own name, so its debts are its own and creditors have no claim against the owners
• •It has perpetual succession, continuing to exist when shareholders die or sell, which makes long-term projects and stable employment possible
• •Shares can be transferred without disturbing the business, which is what makes stock markets possible
• •The company can be prosecuted, fined and, in some jurisdictions, convicted of homicide, without any individual being convicted
• •Directors owe their duties to the company rather than to shareholders directly, a distinction with real consequences in insolvency

The standard case is that it solves a specific coordination problem. Without it, an investor's risk depends on every other investor's wealth, since creditors would pursue whoever could pay, so investing would require investigating your fellow shareholders rather than the business. Limiting liability makes every share identical and therefore tradable, which creates liquid capital markets and lets savings be pooled from millions of people who know nothing about the enterprise. It also permits diversification, since a person can hold twenty companies without twenty unlimited exposures, and it encourages risk-taking on projects with uncertain returns, which is where most innovation happens. The counterargument, made since the nineteenth century, is that it socialises the downside: the company captures the gains and, if things go badly enough, walks away from the losses, which are borne by creditors, employees, and where the harm is environmental or physical, by the public.

The shield is not absolute, and courts and legislatures have cut holes in it. Piercing the corporate veil allows a court to hold shareholders personally liable where the company was a sham, was used to evade an existing obligation, or was a mere facade, though courts do this rarely and reluctantly. Directors face personal liability for wrongful trading, meaning continuing to run up debts when insolvency was inevitable, and for fraudulent trading, breaches of duty and certain health, safety and environmental offences. Banks routinely require personal guarantees from the owners of small companies, which contractually removes the protection for the largest creditor while leaving it in place for suppliers who had no such bargaining power. And limited liability was never intended to cover people harmed by the company who never chose to deal with it, which is the hardest case: a person injured by a subsidiary's negligence may find the subsidiary has no assets and the parent no liability, an issue litigated repeatedly in cases about mining, chemicals and oil operations abroad.

The criticism that has most force concerns groups of companies. A parent can incorporate a subsidiary for a risky activity, take the profits through dividends and management fees, and let the subsidiary fail if the risk materialises, which is limited liability applied not to individual investors but to corporations shielding themselves from their own operations, a use nobody contemplated in 1856. Asbestos, tobacco and offshore drilling litigation all turned partly on this structure. Responses have included statutory duties extending up the chain, human rights due diligence laws in France and Germany requiring parents to police their supply chains, and case law finding that a parent that actually controls a subsidiary's operations may owe a duty of care directly. The underlying tension is unchanged: the same rule that lets ordinary people invest a hundred pounds in a company without risking their house also lets a corporate group organise its affairs so that the people it harms cannot reach the money.

Treating a company as a legal person separate from its owners means shareholders lose only what they invested, which made it possible to pool capital from strangers, trade shares freely and diversify holdings. Britain generalised the arrangement in 1844 and 1856 and the courts confirmed it in 1897. The protection breaks where a company is a sham, where directors trade while insolvent or breach duties, and where banks demand personal guarantees, and the sharpest criticism now concerns corporate groups using subsidiaries to contain risk their parents created.