When talking only about software and CS, I think there is one more way to look at this issue (and has been mentioned multiple times in this comment thread). Le

When talking only about software and CS, I think there is one more way to look at this issue (and has been mentioned multiple times in this comment thread). Let us say the typical software engineer's day is filled up with the following types of tasks: 1. What patio11 describes, although in a different context: "Don't try to make a career out of optimizing the SQL queries to display a preference page on a line of business app at a company that no one has ever heard of." [1] 2. Some kind of algorithmic work (e.g. writing a compiler) 3. Big data, machine learning etc. (consuming the results of algorithms, hence different from 2) 4. Software architecting My view is that after a while, most people want to move from group 1 to one of the other groups. This gives a good way to explain the pursuit of Ph.D. even when it is not economically rational (both in time and money cost) - it is a pursuit of something which is not mundane as long as you make the effort. A teacher of mine said: "You are going to be spending about 40 years in your career. If you take 5 out of that to do a Ph.D. you are not going to reflect on it with regret. And at the end of it you have a Ph.D. too." When you combine this with the possibility that you will be mostly hanging out with elastic minded students as PG once put it and you might actually have the time of your (intellectual) life, the decision doesn't look all that irrational. [1] https://training.kalzumeus.com/newsletters/archive/do-not-end-the-week-with-nothing https://training.kalzumeus.com/newsletters/archive/do-not-en...