Friday 22 August 2008

Working hours

Found a interesting natural experiment on culture at my workplace.

As a manager, my approach to working hours has always been laissez faire. I'm fine with members of my team working whatever hours suit them, as long as commitments to colleagues are kept. Some people come in at 7:30 and wrap up by 4:30. Other come in at 10:00 and work till 7:30. Still other, like me, frequently do a second shift between 9:00 and 10:30 after the kids are in bed. This works just fine.

I recently discovered that my successor at the department I used to run until early 2007 has a different approach. He expects people to be in at 9:00. And he shoos home the laggards who're still working at 5:30. And while the panache and elan that this department had in my time are missing, this approach also works well enough. There isn't one right answer here.

However, the two right answers produce interestingly different selection effects.

My laissez faire approach tends to favour ambitious people willing to work long hours to get ahead. Other things being equal, people who were willing to put in 70 hours a week instead of the typical 50 would achieve more and be rewarded for that achievement. And I did observe a handful of people who were ravenously hungry for success choosing to burn their weekends at work trying to get ahead.

Under a more rigid 9:00 to 5:30 culture the ambitious can't catch up with more talented, knowledgeable or likable colleagues by sheer dint of hard work. Preventing over-long working hours is sort of like a price-setting mechanism in a cartel. People who work too many hours would be "punished" by an external or superior enforcer.

As a result, one would expect people with boundless raw ambition to self-select out of the organization. The culture would increasingly reflect the choices of people with ambition, but who are less willing to make personal sacrifices for the sake of ambition.

Nothing wrong with that. But over time, it does produce an interestingly different culture.

No comments: