Cyril Northcote Parkinson, a British historian, formulated an observation in 1955 that sounds like a joke, yet is confirmed in practice with almost frightening consistency: work expands to fill the time available for its completion.
Give an employee one hour to write a letter, and they will write it in an hour. Give a week for the same task, and it will take a week: additional drafts will emerge, along with unnecessary approvals and pondering over phrasing that no one actually cares about. The volume of real work within the task hasn't changed. What changed is the amount of time spent on it.
Parkinson wrote this about bureaucracy—he observed how the staff of the British Colonial Office kept growing even as the volume of real affairs it handled was declining. He explained the mechanism simply: officials create work for one another to justify their own employment and the expansion of staff.
In modern companies, this manifests slightly differently, but it is easily recognizable: long deadlines do not in themselves guarantee quality—they guarantee that the task will take exactly as much time as it was given. And a meeting scheduled for an hour will find something to fill that hour with, even if there is essentially nothing to discuss.
The practical takeaway is not that you should always set unrealistically short deadlines—that creates another problem: stress and a drop in quality. The takeaway is that a deadline is not a neutral framework, but a factor in itself that shapes the scope of work. Deadlines should be calibrated to the actual complexity of the task, rather than given "with a buffer just in case"—that buffer will almost always be consumed.