Conveniently for Gino and her co-researcher, the bank had asked a similar question in 2007 after launching a new mortgage processing system. It followed its employees' productivity for two-plus years.

Gino & Co. matched the daily productivity scores to Tokyo meteorological data for that time — still expecting that good weather would make workers happy. She had assumed bad weather would leave them grumpy in general and worrying about the weather as well.

But it was not to be.

"We found that an increase in rain correlated with a decrease in the time it took for workers to complete their tasks. To be precise, a one-inch increase in rain was related to a 1.3 percent decrease in worker completion time for each transaction. We also found that low visibility and extreme temperatures also matched periods of high worker productivity. Meanwhile, clear, sunny days correlated with relatively low productivity."

Thus confronted with an unanticipated outcome, the Harvard team probed further.

"In follow-up studies, we examined why this is the case. We found that, on a bad-weather day, people are better at focusing on their work not because the weather makes them grumpy but because they have fewer distracting thoughts about what they might otherwise be doing outside. Indeed, cognitive distractions and error rates were greater on nice days than on bad-weather days."

So maybe employers should be happy when workers call in "sick" on sunny days, or perhaps they could cover the windows and pipe the sounds of a downpour into the workplace on fair weather days. After all, every 1.3 percent counts.