This is a named pattern with documented cases and a sound logic behind it, not an effect anyone has measured. Treat it as a lens that keeps proving useful rather than as a law.
Parkinson was a British historian, and the essay he published in The Economist in November 1955 was satire. It opens with the sentence that made him famous, work expands so as to fill the time available for its completion, presented as a commonplace observation and then dropped. The rest of the piece is about something else, and that something else is what he actually named after himself: the number of people in an administration grows regardless of how much work exists.
His illustration was the Admiralty. In 1914 it employed 2,000 officials. By 1928 it employed 3,569, while the fleet those officials existed to administer had fallen by two thirds and the men in it by a third. Parkinson called the result a magnificent navy on land. His table carries a second comparison that is harder to argue with, because it holds the work almost constant. Over the same fourteen years the men working in the naval dockyards increased by about 9 percent, while the officials and clerks administering those same dockyards increased by 40. He found the same shape at the Colonial Office, whose staff kept rising as the empire it managed disappeared, reaching its greatest size at the moment it was absorbed into the Foreign Office.
The mechanism he proposed is the part worth keeping, because it explains why this happens without anyone intending it. An official who feels overworked has three options: resign, share the work with a colleague, or ask for subordinates. He will not choose the second, because a colleague of equal standing becomes a rival for the post above. He asks for subordinates, and he asks for two rather than one, because a single subordinate would end up dividing the work with him and acquiring the same near-equal footing he just refused to give a colleague. Two subordinates are held in check by each other's ambition. In time each of them feels overworked and asks for two of their own. Meanwhile the new people generate work for one another, papers that must be read, decisions that must be checked, and the work is genuine, and none of it existed before they arrived.
The version of this that reaches most people is about deadlines, and it is true enough: give a task two weeks and it takes two weeks, give it two days and it often takes two days. That is worth knowing when you set a schedule, and it is the smaller half.
The larger half is what happens to a company as it hires. Work does not simply get distributed across more people. It is generated by them. Every new person creates coordination that did not previously exist, and the coordination is real work, and it will fill real days. This is why a team of twelve can feel busier than the team of five it grew from while producing less that a customer would notice.
It also explains a hiring pattern that looks like ambition and is closer to self-defence. A manager under strain rarely asks for a peer, because a peer is a competitor. He asks for reports, and asking for two is safer than asking for one. So headcount requests arrive shaped by internal position rather than by workload, and they are argued for in the language of workload, sincerely.
The useful question when a team asks to grow is not whether they are busy, because they will be. It is what would stop being produced if this role did not exist, and whether anyone outside the team would notice.
Source: Parkinson, Parkinson's Law, The Economist, 19 November 1955, reprinted in his book of the same name. The essay is satire built on selected statistics rather than a study. The 1914 to 1928 window also spans a world war that changed what naval administration involved, which the essay does not address, and the opening claim about work expanding to fill time is asserted as common knowledge rather than demonstrated. The bike shed argument from the same book is the law of triviality.
Parkinson's Law
C. Northcote Parkinson, 1957
The witty original. Read it as a comic essay that happens to be right about organisations rather than as research, which is how its author meant it.
Draw your own card. It does not take long, and it rewards taking your time.