Last week, the media went into overdrive when Marissa Mayer announced that Yahoo was doing away with telecommuting and insisting that employees come into the office and work cheek-to-cheek - or cubicle to cubicle - with their co-workers.
The reactions to the announcement have been fairly typical. On the one hand are articles with pictures of "workers at home" beavering away at desks that seem too clean, and expressing outrage as to how such a decision ignores all the value created by giving people the freedom to work at home or in the neighbourhood coffee shop.
Invariably, these articles include criticisms of the move by other CEOs, such as Richard Branson, who consider them to be out of line with modern workforce practices.
The usual evidence touted by HR types - that firms that embrace workplace flexibility have higher satisfaction ratings and lower turnover - is invariably part of the discussion.
But some articles argue that a ban on telework is no big deal - or that it could be beneficial. These articles point to the importance of face-to-face contact and the serendipitous nature of innovative collaboration.
Many point out that few telecommuters working in major corporations spend the majority of their time in the home offices, as most of the individuals supposedly being more efficient at home are contractors to companies, rather than fully employed with a single company.
But what few of these articles discuss is whether there is any value for a company in having fully employed managers and staff operating in isolation for the majority of their working time.
Of course, there are situations where working away from the office is necessary and efficient. There are distractions to be ignored when concentration is necessary to complete a project or task and being away from the office increases the costs to others of distracting you.
However, the whole point of having individuals working together implies that there are compensating benefits to co-location. So what is the evidence?
First, it turns out that to get the benefit of working effectively with others you need to be quite close to them. Studies in laboratories and technology companies show that the probability of interaction drops to virtually zero at a distance of about 30 metres.
This suggests that if you are on the other side of a building floor (or on another floor altogether), the likelihood that you will interact with someone in those locations is nil on any given day. Hence, one conclusion is that if I, as an employee, am unlikely to interact with these people, why can't I not interact with them from home?
Organisations know this and structure interactions to counter this effect. As strange as it may seem, open-plan offices and all those distractions telecommuters complain about are meant to counter this problem of organisational distance.
While it sounds "modern" to argue for workplace flexibility, successful modern corporations are not just exploiters of knowledge, but explorers and creators of knowledge.
Such creation activities cannot and do not arise from individuals operating independently at a distance.
Second, telecommuting is little more than a form of outsourcing. The best value from removing employees from the corporate prison will arise with those individuals who are least effective or least necessary to increase the value of those around them.
In other words, telecommuters may be very efficient at what they do - indeed, maybe more efficient than a cubicle-bound drone. But what telecommuters are not good at doing is making other people more efficient or innovative.
Indeed, this is the logic behind outsourcing. If a "module" from a value chain of activities can be removed and contracted to someone who specialises in certain tasks and then do this in a market-based transaction then it certainly makes sense to do so. "Output" is simply plugged into the outsourced module.
Outsourcing also is enhanced when there is a clear measurable output and all one is concerned about is making sure that the output meets quality, time and price specifications.
However, what one would want to keep in-house is the sharing of tacit knowledge between employees, which often occurs when employees interact face-to-face. Unlike an outsourcing agreement, it is difficult to assess the market value of these interactions.
A group working together in proximity can meet a corporation's goals in a way that cannot be replicated by aggregating the efforts of individual actors. Unfortunately, these effects are less predictable and harder to manage and, hence, the outputs cannot really be measured easily. This is why managing the "inputs" - the employees - is so important.
For example, I keep my PhD students close to me on a daily basis because the plethora of small interactions I have with them has far more value to them than a single formal meeting.
These points together highlight why being in proximity to your co-workers matters, particularly in knowledge-intensive activities. The burden of proof of the value of an employee in this world is not that you are more efficient, but that your proximity to others makes them more efficient.
Timothy Devinney is professor of strategy at University of Technology, Sydney.
This article first appeared on The Conversation.