Drive: The surprising truth about what motivates us

In Drive: The surprising truth about what motivates us, (animated video, 11 min) from RSAnimate, Daniel Pink, the author of Drive and other popular books, describes some very unexpected results from the recent research on motivation. The old paradigm of higher pay for better performance works - but only for mechanical tasks. If the task requires even the most rudimentary level of cognitive skills (as do most jobs!), higher pay leads to worse performance.
These counter-intuitive results have been replicated in many studies, in multiple countries (USA, rural India, etc) and with different pay grades (including quite substantial 2 months worth of average pay for highest performance in the experiment in India).
So what does motivate people, if not pay? The known non-monetary motivators are: autonomy (you determine what to work on, vs. being told what to do by the boss), mastery (become better at what you do, be recognized by your peers), and purpose (the job must have a purpose which is different from just making a profit for the company). And what about money? If you don't pay people enough, they will not be motivated to work. The best use of money is to pay people enough to get the issue of money off the table, so that people think about work, rather than money.
There are many examples of successful real life use of these non-monetary motivators. Think about Wikipedia, newsgroups, help mailing lists, or open source software projects, such as Linux, Apache, Mozilla Firefox, Perl, Python, R, etc, which are the result of work of thousands of people who are not paid to contribute to these projects.
Google has the 20 percent time tradition, which encourages employees to decide what to work on for 20% of their (paid) time at Google. This tradition led to many successes, including GMail, Google News, and others.
Software company Atlassian (makers of JIRA software, among others) had a tradition where for 1 day every quarter employees worked on company time on projects of their choice, and then presented their results at an informal  meeting. This led to many creative ideas, bug fixes, and other improvements. Following that success, Atlassian switched to Google's model of 20 percent time.

Acknowledgements:
I would like to thank two readers of my earlier posts for pointing me to the Dan Pink video.