Harvard Business School Professor Ethan Bernstein talks about how, by planting researchers on factory lines in China, he found that giving employees more privacy can increase their productivity – a phenomenon he calls the Transparency Paradox.
The Business is a podcast from Harvard Business School that ran through 2015 and took a unique look at the business world through conversations with HBS faculty and entrepreneurs. It has since been replaced by Cold Call, a new podcast that distills the legendary HBS case method into digital form. Subscribe to “Cold Call” on iTunes, and iTunesU or follow us on SoundCloud.
Brian Kenny: Our guest today is Professor Ethan Bernstein of Harvard Business School. He teaches leadership and organizational behavior, in both the MBA and doctoral programs. And he’s the author of an awarding-winning paper called the Transparency Paradox: A Role for Privacy in Organizational Learning and Operational Control. Ethan, thanks for joining us on the Business.
Ethan Bernstein: Thank you Brian.
BK: Why don’t we start with you describing for us what the Transparency Paradox is?
EB: And perhaps the right place to start is the opening line of the article. I start by setting the stage. Organizations’ quest for worker productivity and continuous improvement is fueling a gospel of transparency in the management of organizations. In the name of collaboration, organizations really have striven for transparency in the workplace, literally tearing down walls in an effort to let managers and employees observe each other. As a word, transparency is everywhere, it’s even hit, I think, top 10 lists in various contexts, in terms of usage. But in business schools, we’re interested in performance. So here’s the question: Is all of this transparency resulting in higher performance for individuals, groups, organizations and industries? And I actually find that in some cases, zones of privacy, rather than transparency can be optimum for performance.
BK: What lead you down this path of research? Why did you decide to even look at this?
EB: I was actually interested just in understanding how transparency fueled performance. We talk a lot about social capital. We talk a lot about networks. We talk a lot about being able to share information. We assume that that’s beneficial. But of course anytime that you put human beings in the picture, things can go a number of different ways. And so I actually wanted to understand those dynamics very specifically and very rigorously.
BK: And the way you did it, I think, is kind of cool, it’s very stealth.
EB: If you’re going to study something as hard to measure as social capital, you might want to study some—it in a context where performance is very easily and very rigorously measured, so I ended up in a manufacturing context. And of course at the time large manufacturing - still today - we were going to China. Now if you want to observe people working in their natural setting in China, and you happen to be a professor named Ethan Bernstein, you’re probably not going to be able to do it much yourself and so I actually found five Harvard College undergrads, who had been born and raised in China, had spent at least the first 10 years of their life there, who were willing to come out with me and actually work on the line. So of the 14,000 people or so who worked in this particular facility in China, only three knew that these five individuals were anything other than ordinary workers who’d come and joined the line for the summer.
BK: What were they making in the plant?
EB: The first summer we were there, we actually we there two summers, the first summer it was mobile phones, and the second summer it was mobile 3G data cards. Just to give you a sense of scale, Brian, this facility would’ve made about one in every 25 mobile devices you would’ve found in the world at the time.
BK: What were your observations? What did you find?
EB: So, the first summer we really went out just to, just to observe. And the interesting observations came back, really within the first couple of days. And that was, these individuals who’d been just thrown into the line, as many migrant workers would be, had been trained immediately in two ways of doing things: one, the way of doing things when you’re being watched, and one the way of doing things when you’re not being watched. And that was actually very easy to distinguish, because given this factory floor was really football field wide; transparent, not only could they be seen, but they could see others. So if you saw a pink hat or a yellow hat, signifying various levels of status in the organization walking around, you can modify your behavior accordingly. And that’s exactly what they were doing. Now that finding is not particularly novel, I think we all assume that we operate differently when not being watched. When we put ourselves in the observer shoes, we assume that when we’re not observing, people are slacking off. But what’s interesting here is, they had actually found better ways of doing things, perhaps not surprising given that you were doing the same thing maybe 2400 times a day or at least a shift, but they were doing them better when they weren’t being observed. And so in this case, managing by walking around was actually systematically destroying value for this organization.
BK: You write about the fact that what the workers were actually doing, when they were being watched, was performing. They were giving a performance, to the audience of managers.
EB: One of the questions people have had about this work is ‘is this just a Chinese factory phenomenon or does this translate to all the, business lives of the people who might be listening to this podcast?’ I think, you know I think in my own daily life, when I know I’m being observed, I have a natural desire to meet the expectations of those individuals who are observing me. And I know that if I do something that looks weird to them, not meeting their expectations, then I’m going to attract their attention. When we went back and really understood why it was that workers explained they were doing this, it wasn’t because they were trying to cheat, because that we would think would lead to worse performance. Instead they were really trying to manage the attention of expensive resources around them, whether they were peers on other lines, or managers, or support functions. They just didn’t want to attract attention unless they thought those people could be helpful.
BK: And by managing they were probably using energy to do that, rather than focusing on the tasks that they had in front of them.
EB: Yeah, they were managing up, managing sideways, things that we tell our MBA’s to do all the time. They were just doing it by hiding productive behavior, which seems counterintuitive until you really think about it.
BK: So, when you went back you actually took a group, the second time, and you put curtains around them.
EB: That’s right, so we had planned all sorts of interesting kinds of interventions, things that traditional literature would suggest, this is what you do in this kind of situation to make the organization better. So we had all these plans, we had 32 different lines making the same product; it was a researcher’s dream, run a lab experiment, except in the field where things are actually happening. And we chose four lines to try different interventions with and one of those four lines was a little too close to a control line. And I asked engineering what we should do about that, and engineering said, “Well let’s just put a divider up.” So I said, “Go ahead,” they do these kinds of changes, apparently, all the time on the floor. And the divider started as a curtain bar, about which, apparently, I already have, I already had embeds on the line, so I got to hear all these stories. When the curtain bar went up they, on the line, joked that they could hang their wet laundry and it could dry on there. [Laughs] They were very thankful to engineering for this. And then the curtain went up, and they said, “Oh, the people on the other side of the curtain must have the swine flu.” And then shortly after that, one of the people standing, just by chance, next to an embed, said, “I wish they’d put those curtains up around the entire line, we could be so much more productive if they did that.” And I am a big fan of listening to the field when it tells you something, so it was really cheap to do and we tried it. And, long story short, that curtain intervention on those lines led to 10 to 15% performance improvements over the course of the remaining five month study, which is really, I mean it was surprising to us.
BK: Oliver Burkeman, in the Guardian wrote that, “The open-office is something devised by Satan in the deepest caverns of hell.” So people don’t like these, yet companies still continue to introduce open-office environments and say that it’s going to promote innovation and creativity. Would you agree with that now, or do you think your research kind of is going in the opposite direction?
EB: Up till now, we’ve had some sense, that perhaps people don’t like being in open-offices, but it actually yields higher performance over time. I don’t necessarily have data in open-office setting to suggest specifically an answer to that question. But I do think that the Transparency Paradox, the paradox part of that, has something to say about that. So in the factory floor, the reason it was a paradox was that it became more transparent within the curtain, when the curtain went up. So as an organization we went from hiding and people really—and really not being transparent, even though it looked transparent, to looking less transparent, but actually being more transparent, in terms of the flow of information and the flow of ideas, at least within certain boundaries. I do think that as you expand the boundaries within an office space, you might have the same effect. And so that we pick our boundaries carefully, when we don’t get to pick them, we find a way of hiding within them. And I do think that you may have very similar effects, so rather than getting increased information sharing within an open-office, you might instead have less innovation, less creativity, perhaps even less collaboration, and ultimately less performance. It doesn’t mean it has to be that way, it just means that it could be.
BK: When you think of places that are traditionally open; newsrooms, you know trading room floors; there are places where it does seem to make a lot of sense to have an open environment.
EB: And this goes way back. There’s a lot of literature, for example, in deliberative democracy, to suggest that one of the reasons that courtrooms and jury rooms work the way they do, is because we don’t want them being subject to their initial opinions, when they ultimately make their final decision. The same was true of the Constitutional Congress; you know Jefferson sat outside of the room. Madison was in the room. Madison was very specific about wanting to protect secrecy within the room, within the deliberations; his notes were only released after he was dead. And in those notes is a comment that says, “If it weren’t for our oath of secrecy, we never would’ve gotten anything done.”
BK: Let’s talk about the manager out there who’s listening to this, who says, “My God this is the inmates running the asylum, if we’re not watching what they’re doing, they’re going to cut corners, our quality’s going to—this can’t possibly work in my setting.”
EB: Now recall it, we put up curtains, but I can tell you that they improved their performance by 10 to 15%. So there was something that was transparent, and it was the output, I mean the data showing how they were performing on a minute-by-minute basis, remained extremely transparent. So I would never suggest that what we need to do is take groups of people and isolate them entirely. The question is, for your particular context, what data do you want to be transparent, and what data, in the most general sense of the word, do you not want to be, in order to preserve both collaboration, and innovation, and performance.
BK: What would you say to the worker out there who’s listening to this? Is there some, is there some counselor advice you can give to that person, who’s stuck in this open-office setting and wants an office?
EB: Give your manager a copy of my article. [Laughs] It’s, you know it’s difficult—I go back to the Chinese workers, they ultimately were exercising their agency collectively to hide. Management had said, “You’ll be in an open-office space,” and they said, “Well that’s great, we’re going to preserve our privacy by finding all these nifty ways of hiding things from you, so that you won’t know.” Now that’s not productive for anyone, but it is a worthy reminder, perhaps, the workers and perhaps those worker’s managers, that ultimate agency for transparency doesn’t rest with the folks who are designing their work environments, it rests with the people who are working in them. Because we’re pretty creative individuals as human beings, and if you take away our walls, we’ll find another way to construct them.
BK: It sounds like there’s a lengthy future for the Transparency Paradox.
EB: One I’m always looking for more research sites, so if anyone’s listening and wants to engage in a conversation, I’m always happy to do so.
BK: Professor Ethan Bernstein; thanks for joining us today.
EB: Thank you, Brian. This was great.