Sunday, September 2, 2012

Sharing Less to Learn More: A Radical Rethinking of Productivity

Sometimes management scholars get findings that shake up established truths and make us rethink how firms function. These findings are often directly important to managers as well, because they question common practices that are used with the intention of producing better decision making, higher productivity, and greater competitive strength – but may turn out to have the opposite effect.

Ethan S. Bernstein published a paper "The Transparency Paradox: A Role for Privacy in Organizational Learning and Operational Control" that has just such a finding. The established truth is that transparency in production processes increases productivity because it allows faster learning from others, as problems are immediately seen and effective solutions can immediately be learned by others in the same situation. The idea is to use visibility along a production line to drive a fast learning curve of efficiency increases, and it is a key component of the vaunted Toyota Production System and Total Quality Management practices.

The problem is, his research showed it to be false. In a study design that combined observation of production practices with a controlled experiment, he made some very interesting observations. First, he showed that workers concealed production practices from managers even when production lines were fully visible. Their reasons for doing so was that they had formalized procedures for production steps (also a key quality control practice), and when they found ways to do things faster but still with high quality they preferred to do them under-cover rather than go through the procedure for changing the procedures. But that meant they were worried about getting caught. Second, he showed that the time and effort spent concealing these practices was a form of waste that was especially high when the production lines were visible, because managers could easily see far along the line. Third, he showed that the economic effect of this concealment was big. When some production lines were given privacy (through the kind of curtain that one sees between hospital beds!) while others were left open, the private lines outperformed the transparent ones by 10-15%. That’s a big number in manufacturing efficiency; especially for this firm, which was a contract manufacturer that got all its profits from making goods more efficiently than other contract manufacturers.

So what to do? Clearly, researchers need to find out whether there are conditions that make concealment likely and costly. It is not clear that visible production lines are costly at Toyota and all other firms using them too. If not, then we have to ask why. Managers need to reconsider the role of visibility in manufacturing efficiency. If privacy can give a 10 plus percent improvement in some firms, possibly also theirs, it is a good idea to reassess what they are doing and maybe hang some curtains as an experiment. In a decent-sized manufacturing facility, it takes less than an hour of more efficient production to cover the cost of those curtains!