By retaining a person who has "stagnated" after 3 years (setting aside the ambiguity of what you mean by "stagnate"), you avoid: 1) Three years of lesser perfo

By retaining a person who has "stagnated" after 3 years (setting aside the ambiguity of what you mean by "stagnate"), you avoid: 1) Three years of lesser performance from the new employee as they climb that curve 2) The very real risk that your new hire will not be successful, or if they are successful, will not achieve the same level of performance. Also, I invite you to challenge what you consider "stagnation". It is not possible to continue linear improvement in a given role: good employees achieve mastery, and while they will still be improving that might not be evident to their managers who are looking only at work outputs. If a company does not want employees who are masters at their task, but rather want people to move up or move out (counter-productive as that might be), then it is literally middle management's job to find new roles for successful engineers, and I suppose fire the previously successful engineer if they're not up to the task. Complaining about an engineer "stagnating" in the same role after they've mastered it is just lazy thinking and lazy management.