This editorial was produced in an office, so it may be 13 percent less productive and efficient at its job than if it were written at home. Then again, if it were produced at home, away from the boss’s gaze, it might still be goofing off, collecting glib phrases and sentence fragments without cohering into anything persuasive.

In the latest debate over telecommuting, sparked by Yahoo! Inc.’s announcement that all employees working from home must start showing up at the office, the two camps have staked out their positions. Advocates of working from home cite studies showing that telecommuting benefits employers and employees alike. Opponents extol the benefits that can come only from a spontaneous, collaborative work environment.

Two subtleties undermine each of these views. First, little about telecommuting’s value can be deduced from the current body of research on the subject. Second, Yahoo’s experience with home-bound workers says a lot about Yahoo’s experience with home-bound workers. About the only generalized lesson we can draw is that managing employees well is vital, whether they work in a cramped cubicle or the spare bedroom.

Yahoo insiders say it had become a common view that the company was bogged down with slackers taking advantage of work- from-home arrangements. So Chief Executive Officer Marissa Mayer, hired in July to turn the troubled company around, apparently decided to test that theory by checking how many remote employees were logging into the company’s network, allowing them secure access to Yahoo’s systems. The answer: too few.

How to reconcile Yahoo’s experience with studies showing people are more productive when they work from home?

Most of these reports aren’t studies in any scientific sense. Many are surveys in which employees who work from home self-report on factors such as their productivity and job satisfaction. As such they are vulnerable to the Hawthorne effect, in which subjects change their behavior in response to an experiment. Respondents may overestimate their work output, hoping an employer will permanently adopt telecommuting. In any case, employees are generally poor judges of their productivity.

In addition, even if remote employees produce more in the hours they work at home, there may be related productivity losses back at the office. What about those brainstorming sessions by the water cooler? More prosaically, and realistically, there are the hours spent by colleagues at the office on material to send home and on troubleshooting technology glitches.

Then there are the case studies purporting to show that telecommuting saves money by increasing productivity and retention and by cutting overhead, because companies need less office space. Cisco Systems Inc., for example, claims that telecommuting saved it $277 million a year. Of course, its business is selling communications equipment that enables employees to work remotely. It’s a little like Coca-Cola promoting the virtues of soda sales in schools.

What could help clarify this issue, beyond a healthy dose of common sense, is a randomized trial in which one group within a company works remotely and the other goes to the office. So far, just one such study has been conducted. It is fascinating and has been widely referenced in the current debate, but it has almost nothing to do with the challenges facing Yahoo.

Over nine months, call-center employees for the Chinese travel company CTrip who worked at home proved 13 percent more productive than a control group, based on the objective measure of the number of phone calls they made. The group’s attrition rate, 17 percent, was half that of the control group. CTrip estimated annual savings of $2,000 per telecommuting employee.

What no one mentions, however, is that half of a CTrip employee’s monthly earnings is based on his or her call and order volume. Thus workers had a strong incentive to be especially industrious, and working from home helped: It was quieter than the office, so employees got through calls faster and thus made more of them. They worked more minutes within a mandated shift because, spared a commute, they were less tardy. And they had fewer sick days because they worked even if a little unwell.

For companies such as Yahoo, where output for salaried employees is hard to measure and synergy among at least some workers is vital, the CTrip experience can’t be replicated. Still, for them, telecommuting is not a bad idea per se. It can help attract and retain talent. And it’s well suited for quiet, independent tasks that even collaborative workers do occasionally. The challenge is to establish the right mix between office and remote work and to ensure work from home is productive. That means having managers establish expectations and communicate regularly with their charges.

This was the lapse at Yahoo. It was a failure of management, not telecommuting. That’s why, both inside and outside the company, it’s expected that once the shirkers have been weeded out, working from home will return to Yahoo.

(Bloomberg)