Toyota Motor, the company that the world admires, has been in the news lately for all the wrong reasons. “First of all, I think it’s a very well-run company. So suddenly they were great and now they are rubbish. Neither of those is correct. When they were good they were never as good as we thought. But we were blinded to that because the outcome is so great. And then when they have a stumble, we think everything is rubbish,” says Phil Rosenzweig.
A professor of strategy and international management and the director of the executive MBA programme at the IMD business school in Lausanne, Switzerland, Rosenzweig is also the author of the best-selling book The Halo Effect...and the Eight Other Business Delusions That Deceive Managers. In this interview, he speaks to DNA.
Why is it so difficult to figure out what makes a company so successful?
For starters, because there are so many things going on, both internal and external to the firm, and it is not possible to hold most variables constant and identify the effect of just one.Also because, in business, a firm performance is relative more than absolute, so that even if a company does things well in an absolute sense, it may still fail if rivals do things better.These together make it difficult to identify precise reasons why particular companies perform well at a given time.
Also there is natural tendency to make inferences based on outcomes. That is the halo effect. It is easy to say, “Oh, you know, it is about leadership and customer focus” and so forth, because at a superficial level I infer those things about successful companies and I infer the opposite for the unsuccessful companies. Based on performance, which is seemingly objective and concrete, we often tend to make attributions about other things, such as leadership, culture, execution, and so on.If we are not careful, the things that we claim drive company performance may in fact be attributions based on performance.As an example, consider Starbucks.As long as it is rising in sales and profits, we say that it has a brilliant strategy, a great culture, bold leadership, and is highly innovative.As soon sales begin to slip, we criticise its strategy, claim that the leader became complacent, and assert that it lost its culture.We make attributions based on performance.
You write in your book that lasting business success is largely a delusion. Why do you say that?
It is always possible to select a number of companies that have been successful for many years, then look back and try to offer explanations for that success.Many well-known business books fall into this trap — such as Built to Last by Jim Collins and Jerry Porras.They chose their sample to include companies that had performed well over a long time, and then gathered data, much of which was shaped by the halo effect, to try to explain why.But once the study ended, these companies, picked precisely because of their “enduring performance,” declined sharply — they were no better than the average.Collins and Porras did not explain lasting performance at all.
In fact, if we look over time, the strongest pattern is of temporary success — high performance tends to regress to the industry mean.Very, very few companies avoid this pattern — and just as you think you have found one, wait a few years and you will likely find that it, too, conforms to the pattern of regression.I am not saying that high performance is random, but it is fleeting.
Competition, imitation, sharing of practices, movement of employees, technological innovation - all of these make lasting success unlikely.Yes, we can select some companies that have succeeded for a long time, but that is an artifact of ex-post selection, and therefore a delusion.
What about a company like Toyota, which has suddenly stumbled badly. Is it a good example of the point you have been making about business success being largely a delusion?
First of all, I think it’s a very well-run company. When a company is doing well we exaggerate how good it is and when there is a dip we exaggerate how bad it is. So suddenly they were great and now they are rubbish. Neither of those is correct. When they were good they were never as good as we thought. But we were blinded to that because the outcome is so great. And then when they have a stumble we think everything is rubbish. I write about Dell. Not in my book, but since then. Dell has been a very well-managed company. They have been very good in many ways. But come back to this idea about relative performance. Hewlett-Packard has caught up in some ways.
Apple?
Apple went on a different model but that is strategic choice and has taken certain business away. So as far as Dell’s performance is concerned, are they doing as well? No. Is that an absolute failure or is it a relative failure? It’s a relative failure. And Michael Dell is now scrambling to ask, “How do I get back on top?” It’s not just through better execution. It’s through having to make a strategic choice. But he has to do it under uncertainty.
Can you give us any other example?
Take UBS, the Swiss Bank. Stable. One of the high performers. They have been around longer than Procter & Gamble. Two years ago they had to face a huge loss because of investments made into certain derivatives. So they went from being a profitably run, conservatively managed solid Swiss bank to facing a massive loss.So I was interviewed by Swiss Radio. And they asked, “How do you explain this disaster at UBS?” I said: Be careful. Be careful.98% of the activities at UBS are just as good as they have always been. Retail banking: just the same. Credit cards: just the same. Mortgage banking: just the same. What has happened now is because there has been this enormous catastrophe in treasury, derivatives, financial market meltdown. Now that has created this negative halo when we suddenly think that this whole company is bad. It’s not bad. 98% of this is just the same. Now they jolly well better fix that. And they jolly well shouldn’t have got into that trouble. But let’s not imagine that the whole company has gone from being good to not good.
Now take the case of ICICI Bank. The reason I know ICICI Bank is because when my book came out, they contacted me. And they had bought several hundred copies of the book and distributed it. Frankly, I had never heard of ICICI Bank. I asked them, why do you guys like my book. They said the reason we like your book is because we have been a very successful bank and every time we read about ourselves people think we are brilliant. Now we think we are pretty good, but we know that we are not as good as the press says we are. And if the public wants to think we walk on water that’s their problem. We don’t want to be blinded by our own halo. This was in 2007. And I kind of lost touch with them after that.When I came back again in autumn last year, I went back to meet them and they said that not only was your book good to us when times were good but this last year when times were bad, it reminded us that we are not as bad as the press says we are. So your book has been a nice corrective to not getting carried away to overshooting this way or overshooting that way.
How do you avoid something like that?
The first thing is let me not be blinded by the halo effect. Let me look at aspects that I can truly measure. Even then it is very hard to know. I believe that there are things we can do in terms of competitive analysis, positioning analysis and capability analysis that allow me to make strategic choices that improve my chances of success. It will never guarantee success because performance is more relative than absolute. But it will increase my probabilities of success.
Take us through an example...
I did a lot of work with Schindler in elevators and escalators. Now there are some fundamental questions there. How are we positioned relative to big rivals like Otis, Mitsubishi. Otis is a company that has made many small adjustments in technology. Schindler has made a different strategic approach. They have made a different choice about what part of the world they go in. Some of there markets are rapidly growing. Some are less. How important is China? How important is the Middle East? So they have made certain choices because you cannot be all things everywhere. Then the question is whatever your strategic choice is, how well do you execute it. Do remember the Halo effect, it is tempting to say that you are successful I guess you must execute well. You are less successful you must not execute well. If that is what we mean by execution then that is totally inadequate. So how do we measure execution? What does it mean if you are in the elevator business? It means things like: How do we manage the working capital for the first installation? How long have the parts been lying around before they get installed? When they get installed what is what we call is the first past yield, that means when we first turn it on, does it work or not. Then there are questions about the frequency with which we are called back to fix something and they have metric called mean time between callbacks (MTBC). The bigger my MTBC the better it is. So we are going to measure those things. We are going to measure and then we are going to do it well. Take the case of Bajaj Auto, I was watching these two wheelers roll off the line and the last guy turns the key and they all start up. I said, how often does it not start up?
They say it’s very, very, low now.So these are examples of execution, we don’t infer it based on outcome, we measure it rigorously. We make a choice. We are going to be active here, not there. We are going to perform this and not that. Then I am going to execute as well as I can. I think that’s what makes these companies high performers. Will they therefore be high performers forever? Of course not. The more dynamic the industry, the more faster the regression.
It’s been three years since you published The Halo Effect. Do you see a lesser tendency among management gurus to ascribe success of companies to particular strategies since then?
The short answer is no. I got many replies from readers and colleagues who said that it was eye opening and it changed the way that I am doing things. I got a lot of replies from people in business from fields I did not know well, safety engineering, risk management, pharmaceuticals etc. They liked the book and I thought what is the commonality among these people? The commonality was that these were people who appreciate valid data used to reach valid conclusions. And those kind of people really liked what I wrote. But I must say that this is a small minority.
What about Jim Collins? How did he react to your book?
He never answered my emails. He has never admitted to reading my book. He says he is far too busy. A few people have said to me that he understood some of the issues and has tried to do things a bit differently in his new book How the Mighty Fall.
We recently interviewed Collins and asked him “Is there anything like lasting business success or is it all a delusion?” He asked us “What year was Procter and Gamble founded?” “Sometime in the 1800s,” we replied. “1837. Is Procter and Gamble a successful company today?” he asked. We said “Yes.” “QED,” he replied.
If you take a large enough sample size there will be through totally random probabilities ups and downs. The next question is, “Is it a success?” If success is defined as survival, they are successful. If success is judged by outperforming others in the industry, they might be a success, but they may be the exception that proves the rule. Over long periods of time, the fundamental rule is of rising and falling, ebbing and flowing. I have never said and I will never say that business success is purely random. It is not luck. It is a matter of making choices and but they are choiceswe make under uncertainty. My fundamental difference with Collins is that he has the view and he states this explicitly, that there is a blueprint to success. And if you do these things you will be successful to the certainty of physics. That is a fundamental misconception. It does a great disservice because in physics or in chemistry or in natural sciences, the phenomena that we study tend to be ones where performance is absolute.
But what about Procter and Gamble?
Procter & Gamble is an interesting
example. Soap has been around for a long time and nobody is going to invest a very different soap anytime soon. And brand preferences are very stable. And compare it to one of these guys (showing his iPhone) where what you had five years ago is very different to what you have now. Again when Collins says 1834, Q.E.D., well yes, if you are Gillette razors, Nestle Chocolates or Kellogg’s cornflakes, those have very long lives. Nestle, they make certain chocolates that have been around here forever. Nobody is going to invent a very different chocolate tomorrow. By the way, they continuously adjust certain things. Next time you look at a bottle of Heineken, looks at what it says. Gold Medal Paris Fair 1902. What are they saying? They are saying, this is the same beer for more than a 100 years. This (waving the iPhone) is very different. Be very careful when people give examples from Procter & Gamble and take that to apply it to Dell or Hewellet Packard.
Now there are lot of good things that Collins writes. Who can be against having good values, caring about your people, caring about your customers? Who could ever argue about that? I prefer humility over arrogance. I mean, who could every argue against that? But fundamentally that’s not what he is saying. What he is saying that you do these things you will predictably be successful and that is a huge disservice. And then he writes this book How the Mighty Fall and there again it’s classic Collins. It is sample selection based on outcome. So I take a bunch of companies that fell and then I look back and say that you know what, they all suffered from hubris. Now hubris is a very interesting word. Hubris is always inferred retrospectively. In fact, almost by definition because what is hubris? If you go back to Greek mythology, hubris is excessive pride that is punished by the gods. When we start saying that you suffered from hubris, we always do it through the lens of retrospection and what he has not yet recognised is that valid research does not begin by selecting samples based on outcome and then gathering data through the lens of those outcomes.
So Collins basically makes management look like physics, which it is not?
It’s not E=MC2. Don’t say it’s physics.It is total nonsense. And that to me is a disservice. I will tell you a little story. I visited once a manager, an executive of a company in The Netherlands. So I went into this man’s office. The desk was very neat. Not a piece of paper in sight. On the bookshelf there were two books. There was the company phone book and there was Jim Collins’ Good to Great. I said that is interesting. So let’s do a little mental experiment, I told myself. Let us imagine that this is a guy who does not read business books except one. He reads Good to Great. What I hypothesised is that if he reads only Good to Great, he will be a little bit better off but a little bit worse off. I will tell you that he will be a little bit better off. Why? Because Good to Great says have focus, be humble, care about your people, listen to your customers. I would say that on the margin there is more good than bad. Of course, the other thing it does is that it says that there is formulaic predictability that I think is incorrect.
Tell us something about the EMBA programme of which you are the director?
There are of course many executive MBA programmes out there, but most of them operate on a very different model than ours. And the reason for that is that most of the other ones tend to be based in places like London, New York, Los Angeles, Singapore or Paris where there is a large population nearby. So most of those programmes are structured in a way where the executives will come, maybe for a long weekend, once a month for 18 months. When you see where our students come from, we can’t ask them to come to Switzerland for a long weekend, once a month. It is just too far for them to come. So we have taken a different model. We have divided our programme into two stages — the first stage is called the foundation stage and that is two times a year; five weeks’ residential in Switzerland. And then they come to the masters years — the EMBA year, which is the year I direct. The idea there is that we will come together four times a year and but the rest of the year is really handled on a distant basis. The central pillar is we have them do in-depth work applied to their own company. We have them do a strategy assignment, a marketing assignment, listen to customers, a finance assignment, an organisation assignment, how are they organised. Most of them are sponsored by their companies and these company-specific assignments are very good ways to learn more about the company and make recommendations for the benefit of the company. And many of the students come away and say this has been hugely beneficial for my company and it has also been beneficial for me because I normally don’t work in finance or with customer or with strategy. But the assignment has given me exposure to those issues and given me exposure to the senior managers.
Comments