Monday, April 30, 2012

Do Reputations Matter?

The Economist's Schumpeter column recently contained a commentary on corporate reputations and reputation consultants, giving three reasons why firms should worry less about their reputation. First, it is unclear exactly what a reputation is: firms do so many things that people have opinions on that there isn’t a good way to add them up to a single object and call it reputation. Even if they can be added up, a manager would need to split them back up in order to make a specific decision. Second, it is not clear that reputations are any good. While we know many high reputation firms that are also profitable, low-reputation firms are too. For example, the tobacco industry is not widely admired, even by its customers, but with low costs and a sales pitch that includes addiction, it is highly profitable. Third, it is not clear that attention and actions to improve reputations actually have that effect. Firms are rewarded for products that perform reliably, for good service experiences, and for service and care in operations. These produce reputation, but are so obviously good things that a manager ought to see their merit without having reputations in mind.

So far the argument from the Economist. Can research help justify all the attention toward reputations? Actually there are some reasons to pay attention to reputations, but they differ from what you might hear from a PR firm. First, reputations spread. Problems in one firm will reduce reputations of other firms that are similar to it, even if they are innocent. I have shown this effect in research with coauthors Stefan Jonsson and Takako Fujiwara-Greve. This has some surprising implications. I wrote earlier about how the Carnival CEO kept quiet when their subsidiary Costa Cruises dealt with the aftermath of the Costa Concordia shipwreck. Surprising as this stance may seem, it is actually an effective way to avoid the negative reputation from spreading, because the connections between these firms are not so well known now, but would become obvious if Carnival made comments on the Costa Concordia shipwreck.

Second, reputations linger, and continue to hurt the firm even after they have recovered from the problem that caused the reputation drop. What’s more, we know less about how quickly a reputation recovers, and why, than we do about how actual quality covers. On the latter, there is already good evidence that reputation-harming events spur learning and improvement, as in the work by Haunschild and Rhee cited below. One could calculate the cost of events such as car recalls based on their findings except for one thing: they show the actual quality changes, but in addition to that there is a (probably lagging) change in the perceived quality. We would all like to know about that effect, because it could be one more reason for firms to pay close attention to quality, reliability, and safety.

Haunschild, P. R. & Rhee, M. 2004. The role of volition in organizational learning: The case of automotiveproduct recalls. Management Science, 50(11): 1545-1560.
Jonsson, S., Greve, H. R., & Fujiwara-Greve, T. 2009. Undeserved loss: The spread of legitimacy loss toinnocent organizations in response to reported deviance. Administrative Science Quarterly, 56(June): 195-228.
Schumpeter: What's in a Name? The Economist, April 21st 2012: 73.

Tuesday, April 24, 2012

Olympus Cover-up: Beware of Unstable Power

Last week we were treated to the spectacle of Olympus Corp. having its first stockholder's meeting after CEO Michael Woodford lost his job after alerting the firm’s board of directors of possible accounting irregularities. As a shareholder in Olympus, Mr. Woodford had the right to ask questions, and took the opportunity to ask for an explanation of why he was fired. This question was not answered by the company management, which is significant for two reasons. First, failure to answer question from shareholders can make a meeting invalid, meaning another one would have to be held (the odds of that happening are low). Second, it would be interesting to hear a public explanation of why a CEO was abruptly fired when informing the board of potential irregularities. It is after all what CEOs are supposed to do.

Observers of these events are not much in doubt about the reasons for the firing. The chairman of the board at the time, Tsuyoshi Kikukawa, saw the information about the irregularities as a threat to his position and legacy as an Olympus manager, and had the board fire Mr. Woodford as part of a cover-up. This was a hugely risky decision. As a cover-up it had minimal chances of succeeding because information about accounting irregularities was already circulating in the press, and a fired Mr. Woodford would be likely to go public with the charges, as he indeed did. As a way to protect oneself, it was probably the most conspicuous act Mr. Kikukawa could have chosen to interest and annoy prosecutors. The Securities and Exchange Commission in Japan investigated the irregularities and handed the report over Tokyo prosecutors, who promptly arrested Mr. Kikukawa and other individuals involved in the irregularities. If the charges are proven in trial, each could get a prison term of up to 15 years.

It is difficult to know what goes through the mind of someone who makes such an obviously bad decision as Mr. Kikukawa (and the Olympus board that supported him). An interesting clue comes from the fact that as the Chairman of the board he had a very powerful position in the company, but Mr. Woodford’s investigation of the accounting irregularities made this position unstable. In a study by Jordan and coauthors, this combination of high power and low stability has been found to lead to more risky decisions than either stable and high power, or low and unstable power. This is because having something to lose is stressful, which can lead to risk taking beyond the level of those who have stable high power, or those who have low but possibly increasing power. In short, someone powerful who is trying to control a potentially serious threat is likely to make decisions that look puzzling to others.

So the prosecutors have made their move, but this has not prevented Olympus from putting in place management much like the previous one, and a board heavily dominated by banks. What about transparency, rethinking of the strategy, or a greater voice to the shareholders? That does not seem to be on the agenda. But the same paper by Jordan and coauthors also found one more thing: those who have low and stable power also take high risks. So maybe a rebellion of shareholders is brewing?

Jordan, J.J., N. Sivanathan, and A. D. Galinsky. 2012. Something to Lose andNothing to Gain: The Role of Stress in the Interactive Effect of Power andStability on Risk Taking. Administrative Science Quarterly, 56(4): 530-558.

Friday, April 20, 2012

Disney Studio Chairman Out: Starting a Turnaround?


News just got in that Disney Studio chairman Rich Ross is stepping down, with no replacement announced. Rich Ross had a successful run at other Disney businesses including most recently its cable network Disney Channel, so his short tenure at Disney Studios (less than three years) comes as somewhat of a surprise. Or maybe it doesn't: Disney Studios are looking at a series of very disappointing movie releases, including the recently released "John Carter", which is already proving to be a major flop. The studio also has an unclear set of policies on what kind of movies to sponsor, and an unusual and complicated production structure that draws movies from a range of affiliate studies such as Pixar, Dreamworks, and recently acquires Marvel. In short, Disney Studios is a turnaround candidate.

The first page in the turnaround playbook is to replace the top management—usually the CEO (Disney is replacing the chairman instead). The substantive logic is that the CEO is too committed to the current strategy and may lack the capabilities for the new one; the symbolic logic is that everyone inside and outside the firm will know that the turnaround is serious if a new CEO is in place. The substantive part of this logic is slightly puzzling when the CEO has served for a short time, because the performance and even many actions will still be affected by the predecessor's decisions. So does it work? The obvious way to check is to see if CEO replacement in a turnaround situation improves performance or not.

A recent study by Chen and Hambrick finds that the first page of the turnaround playbook is too simple: Just replacing the CEO has no effect on the improvement of firms in turnaround situations. Instead, what is needed to improve the performance is to examine whether the current CEO has capabilities that match those of the firm and its strategic context, and whether a possible new CEO might have better fit of capabilities. Boards of directors (or corporate owners, in Disney’s case) need to think differently on two dimensions, then. First, there is no leverage in simply making a replacement. Second, replacement is not a matter of getting someone better than the current CEO on an absolute scale; it is about specific capabilities. Finding out what those capabilities are calls for serious analysis; it also suggests that replacing a top manager without identifying a successor at the same time is a gamble. But then, in the movie industry pretty much everything is a gamble, so maybe Walt Disney Co. CEO Robert Iger is still comfortable with this decision.

Orden, Erica and Ethan Smith. 2012. Disney Studio's Chief is Out. Wall Street Journal, April 20 2012.
Guoli, C., & Hambrick, D. C. (2012). CEO Replacement inTurnaround Situations: Executive (Mis)Fit and Its Performance Implications. Organization Science, 23(1), 225-243.

Tuesday, April 10, 2012

Instagram for $1 Billion: What is it Really Worth?

Facebook has just acquired the maker of the iPhone application Instagram, which lets users do simple photo editing and sharing with friends, including social-network functions such as “like” and comments. The application itself is in no way complex enough to explain the $1 Billion price (even the valuation of the company on Feb 2., $30 million, would be a rich price for such application). The earnings are not either: There are no revenues at present because the firm has concentrated on building a user base, and has not launched any revenue model, such as advertising or other commercial tie-ins.

So what is driving the price? One, the user count is increasing very rapidly, going from 15 million in early December to 27 million now. Well, it is probably more because I am writing 36 hours after the original Wall Street Journal article. Two, the application has high engagement, meaning that users spend a lot of time with it. There are probably some other reasons as well, such as the speculation that Facebook felt a need to defensively grab this firm before someone else did, but those two explanations sum up a lot of the appeal. What we have is a valuation based on the network effect: users find an application much more valuable because others (especially their friends) also have it, and this network effect makes them reluctant to change to a competing application that offers the same functions, or even a better one. Once the network effect is established, Instagram may be able to reduce the user value, such as by devoting part of the screen to advertising, and the users will still remain loyal.

Or will they? Under what conditions can the network effect unravel? The answer is in theory surprisingly simple, but hard to do in practice. A business based on the network effect can be taken apart through knowledge of how diffusion processes work. First, make a better product. (This could be a problem in the Instagram case: possibly the simple photo editing and sharing is all its users need or want.) Next, generate adoption of your new product by technologically curious "innovators". This is easy to do; a certain proportion of the population like to try out new things to see if they are better than what they currently have. Among these innovators you will find some opinion leaders that others pay attention to: make sure to announce their adoption if you can, and give them tools to invite their friends. Social network sites do this through their “friend” or “link” contact requests, and any application that tries to break into a network-effects market needs such a function. If the business you are competing with is well established, find a segment where it is weak and start there. If your service is really good, it will naturally start eating into the next segments of the population from there: The early adopters, who are influenced by the first innovators, the early majority, who are the friends of the early adopters, and the late majority, who come on board when something is so big that everyone is telling them about it. Even if your service is not that much better you may be able to find ways of rewarding users to stay (usually through neat product features) while waiting for the other to make some misstep, like putting too much advertising on the screen.

Businesses built on network effects have customer loyalty that makes them valuable beyond what a simple examination of the product or the revenues would suggest. But they are not isolated from competition and they are not licenses to print money. A business built on network effects can be beaten by another business that also uses network effects, plus good product design.

Raice, Shayndi & Spencer E. Ante. 2012. Insta-Rich: $1 Billion for Instagram. Wall Street Journal, April 9 2012.

PS: Here is an interactive look at the numbers behind the acquisition:

Wednesday, April 4, 2012

The Band AKB48 and Organizational Capabilities

If you do not live in Asia, you may not know the band AKB48 very well, but chances are that you will hear about them soon. They are a Japanese all-girl pop music band (if the name applies) that have something between 90 and 100-plus members depending on what level of affiliation you count. They have full members, affiliate members, and so on. They perform very frequently in different configurations, have fan clubs for individual band members (yes, the fans can keep track), and even votes for the most popular member. I will refrain from commenting on the music, but the performances involve singing and dancing.

AKB48 brings the idea of a band (if the name applies) as close to an organization as it can possibly get, and they are worth studying for that reason alone. The band is a franchise organization with a carefully cultivated brand, a strong marketing message, standardization of the product and even (because they perform in different configurations) the ability to deliver similar products (well, songs) in many places at once with good reliability. Remarkably, you could go to an AKB48 concert and know what kind of show to expect, even if you did not know who was going to deliver it. That is what good franchise organizations strive to accomplish (of course, I mean no comparison with franchises like McDonald’s on other dimensions).

The band members come in tiers depending on their experience and skills, and younger ones are carefully trained from being rather unpolished performers to reaching the first-tier AKB48 performer level. All this happens in full view of the fans, who often pick favorites among the younger members and follow their career as they go up the tiers of membership. AKB48 have even developed to the point of getting a fairly systematic career system that lets the most popular performers "graduate" into advertisement deals, television acting, and solo record contracts, while new ones enter the organization and start the same career path again. The franchise form of organization is known for its ability to deliver service reliably, in multiple locations, and over long time, and is also quite good at building up capabilities through experience. As a result, we often eat in franchises or get other services done there such as photocopying and haircuts, but even I had not anticipated seeing a franchise band.

Kenneth Maxwell. 2012. AKB48: World Domination, Phase II? Wall Street Journal Asia, April 4.