Chronicler of corporate culture
By Cat Hofacker|April 2024
Amy Edmondson, organizational psychologist and professor at Harvard Business School
★ Click at right to open this story’s Knowledge Guide entries.
★ Click star at right to open Knowledge Guide.
Dave Calhoun choked up slightly as he addressed employees at Boeing’s factory in Renton, Washington. “Every detail matters,” he said in a video shared by Boeing in the days after the door plug blew out of a 737 MAX 9 during an Alaska Airlines flight. To organizational psychologist Amy Edmondson, the CEO’s tone was a marked change for Boeing compared to the months following the MAX crashes five years ago. This could be evidence that the internal culture is shifting. I reached Edmondson at her home office in Cambridge to talk about her specialty of organizational psychology and how it might apply to understanding Boeing’s situation. We spoke before Boeing announced that Calhoun will depart at the end of the year. Here is our video conversation, compressed and lightly edited.
Q: Why should someone in aerospace care about organizational psychology?
A: Early on in my career, I became fascinated by how smart people in great, storied organizations were unable to help their organizations make the changes that they and everybody else could see needed to be made. Take General Motors in the mid- and late ’80s: Everybody knew that the customers were looking for higher-quality, less expensive, more reliable cars, and Honda and Toyota were starting to really break into the U.S. market. It wasn’t just the person on the street who saw that, General Motors saw it, too, but they struggled to make those changes. So I became interested in the phenomenon of the learning organization, an organization that adapts and innovates and learns as needed in a changing world. I became interested in the human elements of that, the managerial elements: What do you need to do to create a culture where people can experiment and try new things and make changes and ultimately shift the Queen Mary in small ways.
Q: What is it about the workplace that makes humans interact differently compared to other environments?
A: We know a lot about groups and group dynamics. The classic social cycle research largely pertains to groups of peers, whereas in a work environment, most groups have hierarchy, a manager or a team leader. That’s been one of the phenomena that I’ve been most interested in, the sort of natural responses that people have to hierarchy that often lead to counterproductive responses. Hierarchy can lead people to hold back an idea, to not speak up when they have a concern, to not disagree with the boss and all sorts of other things. There’s either an implicit or explicit belief that that person has some control over valued resources and rewards for me the employee, and we intuitively understand that the costs of speaking up — the costs of dissent — are borne by the individual, whereas the benefits are shared by the group, meaning the company. So if I speak up and say, “This technology doesn’t work,” I may suddenly be unpopular; I might lose my job. It rarely happens, but there’s that kind of belief about it. Whereas if I stay silent and hold back and just go along, I won’t pay any penalty, and I don’t feel as acutely the cost that that may end up creating for the customer or society.
Q: That sounds similar to how CEO Dave Calhoun described Boeing’s “culture problem” on a podcast last year: People were not being honest with each other about the amount of time it would take to accomplish certain tasks, like returning the 737 MAX to service.
A: I think he’s right in the sense that I do not have any evidence of or inclination to believe that Boeing doesn’t care about safety. Obviously, everybody cares about safety and everybody cares about profits, but in the moments that matter, they’re not putting two and two together in a fully logical way. I write about this story in my book, “The Right Kind of Wrong” — I turned the book in the summer of ’22, so this was before the plug door incident this year — that the 737 MAX failures are what I would call complex failures, meaning they aren’t single-cause failures where there’s just one lever that was pulled or one decision or one error made. They’re multicausal, and we can trace that multicausality back to the McDonnell Douglas-Boeing merger. It changed the culture; it changed the ethos of who’s in charge and what matters. The executives and headquarters were relocated to Chicago rather than Seattle, for instance.
The merger was finalized in 1997, and the headquarters was moved in 2001. — CH
This confluence of factors lines up to create more than the sum of the parts in a problematic way. Calhoun strikes me as a very thoughtful person and not as someone who would have ever wanted any of this to happen under his watch, nor would anyone else at Boeing be willing to trade off safety for profits explicitly. It’s more that moment to moment you are making that trade-off, and there are multiple pressures reinforcing that, including the unexpected introduction of a new Airbus.
She’s referring to the A320neo, announced in 2010. — CH
That basically put a hurry on, “OK, we need a plane to compete with that. We know we can’t get one totally from scratch in that time, so we’ll extend our fabulous 737.” You know the rest. So you can see how you can make lots of little reasonable decisions in their own right, but then when you follow the whole trail of reasonable decisions, you end up in a place that doesn’t make sense.
Q: So when there is a tragedy like the MAX crashes or NASA losing the Challenger and Columbia shuttles, what are the indicators that it is a one-off event versus an indicator of a wider problem with the company culture?
A: I don’t have a perfect answer, but I have a partial answer. The question I would ask is, “What’s the evidence here that there’s a problematic culture?” A problematic culture is one in which people don’t have the confidence that they should always speak up, even if you’re going to be wrong. The other thing I would look for is evidence that there are systems in place and routines in place to routinely anticipate the things that could go wrong, rather than just go about our business. It’s one thing if you’re running an automotive assembly line and something goes wrong. You’re going to lose $6,000 on the margin of that vehicle because you’ll have to fix it, and you’ve lost that opportunity to sell it today. If you are in an organization with far more immediate life and death consequences to your business, then you must have a higher level of vigilance and alertness. I’d be looking for whether there is normalized deviance, the term coined by Diane Vaughan, who wrote the definitive book about the Challenger launch situation. That is where an organization has these rules and protocols, but people don’t really take them seriously because nothing ever happens. Efficiency matters, of course, but when you get the signal that efficiency matters more than curiosity, you might worry.
Q: In the big picture, is Boeing an outlier, or do many companies struggle with their culture?
A: I don’t have systematic or representative samples that would allow me to say yes on average, but I will say from my work with a lot of companies in a lot of different industries, it is extraordinarily rare to see something that palpably feels like a learning culture of transparency and curiosity and honesty and a willingness to confront itself in the mirror and say, “OK, where are we falling short?” A marker is when the conversations around consequential decisions are what I would call high-quality conversations: Everybody’s leaning in to participate, nobody’s on their phone, they’re either listening intently or speaking thoughtfully at any given moment. They might even be watching the dynamics unfold in case you see some people who seem to be disturbed but not speaking up. There’s a healthy mix of statements and genuine questions. “What are we missing?” or “Can you walk me through that so I understand it better?” Finally, a high-quality conversation is one in which you have a palpable sense of learning and making progress, like we’re all getting smarter by virtue of having participated in it. So with those criteria of true excellence in terms of management, those organizations are pretty rare. Another way to put that is in most organizations that I’ve studied, the way they operate operationally and interpersonally is not where it should be given our volatile, uncertain, complex, ambiguous world. It’s almost like it would work fine maybe in the Industrial Era, where we had a longer line of sight, where we had more consistency, more certainty about what would happen this quarter, this year, these next five years.
Q: That seems to suggest that there needs to be a shift on the management side to put more emphasis on flexibility or adapting quickly.
A: Yes, learning flexibility, agility, honesty. The quicker we know the truth about anything, the better able we are to solve that problem — and in most organizations, it’s not so quick. The senior executives are fairly happy because they’re not hearing any bad news, the folks in the middle are kind of somewhere in the middle and the front line is going, “Oh no.”
Q: When you look at Boeing’s response to the MAX crashes in 2018 and 2019 versus the Alaska Airlines plug door in January, do you see evidence that they’re trying to change?
A: I do. Drawing from publicly available articles, Calhoun’s remarks, there’s a very striking contrast between the reactions. With the 737 MAX, there was a sense that they were trying to throw the blame. That’s not great in any way, but they knew they hadn’t fully disclosed the implications of the MCAS [Maneuvering Characteristics Augmentation System] and the lack of training requirements and so forth.
Prior to the MAX crashes, the MCAS software was not mentioned in the 737 MAX pilot manual, and an investigation by the U.S. House Committee of Transportation found that Boeing did not disclose that MCAS would push a plane’s nose down if readings from a single angle of attack sensor indicated the aircraft was approaching a stall, among other details. — CH
Today, I think the reaction has been much more “OK, we’re looking into it. We have to understand it. We have to take accountability — not for everything, but for our part, and our part certainly has mattered.” I don’t have any way to comment on whether this is getting fixed or not, but it looks positive outwardly.
Q: Calhoun said in the February earnings call that Boeing would “encourage and reward employees for speaking up.” If the prior culture had not encouraged that, how hard is it to make the shift?
A: It’s difficult, but not as difficult as people think. It’s about creating a culture of psychological safety, a sense of permission to raise concerns, to speak up. And in a way, he just granted permission. So part of whether or not this will work is how people actually respond, but he did just invite it, right? I have seen some pretty quick changes from doing no more than issuing the invitation and explaining why. For Calhoun and for Boeing, it’s pretty obvious why: “We don’t want this to happen anymore, so we need you.” And they’re saying, “We will reward, not punish,” which overrides the natural intuition of people to just hold back. That’s good, and I think it is doable, particularly given that I don’t think there’s a single Boeing employee who doesn’t want to do better. So the motivation is there, the permission is there; now, they’ve got to develop some new habits and help managers at every level to actually do the reward part. The one caveat I would add is that from what I’ve read, there were people previously speaking up at Boeing, and they were not heard, they were not listened to, and in some cases, they were branded troublemakers or whistleblowers.
Q: You’re referring to prior to the MAX crashes, or earlier incidents?
A: Even going back a little further; I remember reading about the Dreamliner production where parts were being left in the planes and so forth.
A 2019 New York Times investigation of Boeing’s 787 South Carolina plant quoted employees and internal documents describing metal slivers and other debris left near the cockpit wiring and other areas of aircraft. — CH
So those employees were contributing to a speaking up culture. The other part of psychological safety is that you need a listening culture. I think one of the things that turns people off about the idea of psychological safety or speaking up is that everyone will just turn into a big whiner — “this is wrong and that’s wrong” — and then they will sit back and hope someone else magically fixes it for them. Actually, the fixes are going to be collaborative, that this is a team problem-solving opportunity. I speak up because I care, but I also know that I’m at least partially on the hook to work with you and others on a solution. It’s really a fuller participation at work. Instead of sitting back and doing my job, I’m all in. I want to point out what isn’t working, help fix it, be a good team member, care about the customer. I want to feel good about myself at the end of the day because I can leave and go home having done a really impressive day of work with my colleagues.
Q: Based on what you’ve described, it sounds like everyone has a role in maintaining the company culture, but creating the culture or changing it needs to come from the top and be reinforced in the middle.
A: It’s best if it comes from the top because the top is the one that has influence over everybody, but don’t wait for the top. If you are an employee at company X, a manager or team leader or a supervisor, I promise that you can make a difference. You can’t make as big a difference as the CEO, but you can make a difference where you are. If you show up more passionate and curious and interested and honest, you’ll influence your colleagues, and if you’re a manager and you’re doing that, you’ll have an even bigger influence.
Q: But if that kind of initiative isn’t rewarded, isn’t it discouraging?
A: It can be discouraging, but I’ve also seen, in almost every organization: pockets of excellence. And it’s so much better to work in one of them, right? So let’s say the CEO isn’t so great and those folks over there aren’t so great, but this branch or this unit is actually a great team, and that matters. It’d be better to have the whole company be that, but there’s almost always a variability across groups. Even in organizations with fairly strong cultures, you’ll have better and worse [pockets], and it’s often attributable to local leadership. So while it’s helpful to have the right message from the top down, you can also just envision the possibility that an organization just gets better and better because more and more units get to be good.
Q: We’ve talked about rebuilding internal trust, but what about the external trust with airlines and the public?
A: I’m not sure how much that is related to changing the internal culture, but it does matter. With something like Boeing that does touch all of us or most of us in that we are customers, the visibility of Boeing’s troubles is greater than most. I think Calhoun’s intuitions are very good to understand that we need to not only rebuild the trust from engineers, we have to rebuild the trust from the public and that those external mail messages matter. I know the statements about accountability and transparency are words, but they’re not easy. We don’t always hear leaders saying things like that.
Q: What are the external signs that a culture shift could be occurring?
A: It’s hard to say anything numerical about time, but often what you see is a few visible projects or tasks forces that are addressing various issues that really matter, whether it be a drop in market share or a kind of an engineering problem somewhere. You’ve also got maybe a handful of select and focused groups doing some very fast work to address some crucial issues. Culture change is slower, but it doesn’t take long for new messaging to get around. And if new messaging is communicated and then people have help applying it or putting it to use in their own realm, in their own work, things can turn around pretty quickly. It’s what I call a breakthrough follow-through model: There’s some new messaging that really wakes me up, and then there’s some structure and support to help me actually put that to work in what I do.
Q: There are always setbacks when attempting to make a long-term change, so how should a company react?
A: Both in the beginning and all along, just acknowledge that “This is hard, we won’t transform overnight,” but this is how we do it, right? Every time we encounter a bump in the road and we speak honestly about it and we figure out what we need to do to solve this problem and face forward again, we’re doing what we’re supposed to do. The reaction must be learning oriented. That’s probably the best description I can give. Ask “What happened and why? What do we try next?” It’s not casting blame. Of course, now and then you discover that someone engaged in an active sabotage, and it’s unacceptable. In those cases, swift action on actual blameworthy acts is as important as appreciation for speaking up with ordinary challenges and problems. In fact, I think it’s as if not more important to punish actual blameworthy actions as it is to reward candor and speaking up and smart risks. If you don’t have both, there’s no real foundation and no rigor.
