Saturday, October 4, 2014

Path Breaking, Alliance Breaking: How New Technology Complicates Auto Manufacturing

The Paris Auto show is usually a time to celebrate new models, show off prototypes, and generally be in a good mood about what each car maker has done, and even what the industry has done. It is hard to be really upset during such a "celebrate and show off" occasion. Yet this year the CEOs of Daimler and Renault-Nissan attacked Toyota’s plans to launch a new line of environmentally friendly cars based on fuel cell (hydrogen) technology, with Ghosn of Renault-Nissan asking "How are you going to market these cars?"

Funny question, that one. One would think that marketing a car would be Toyota’s problem, and that Renault-Nissan would not be worried about the marketing being difficult, or even impossible. Toyota’s loss would be Renault-Nissan’s gain, right? Toyota responded that Toyota had introduced new technologies before, an unsubtle reminder that Toyota came up with the Prius while European car makers thought that diesel was the path toward environmentally friendly driving.

To see the reason for the conflict, one has to realize that auto makers that are not alliance partners, like Toyota, Daimler, and Renault-Nissan, sometimes act together in order to advance technologies that they all depend on. Through legal requirements they are being pushed to produce zero-emission vehicles, especially in the important California market, which they are currently addressing through making electrical cars. But electrical cars are hard to market, in large part because there are too few recharging (not refueling) stations around. In order for electric cars to succeed, the car makers are strongly dependent on the state of California and others to make good recharging networks for them. 

For a while, Toyota has been playing along with the others. Although their skepticism of electric cars is well known, they finally made a rechargeable version of the Prius (but not until the third generation model…), so one can now drive a Prius without ever starting its gasoline if buying the plug-in version. Yes, it is a special version; a regular Prius cannot be recharged, only refueled. But Toyota isn’t satisfied, and it isn’t particularly concerned about hurt feelings among other car makers. It thinks fuel cells are the way forward, even though that technology is difficult to master and also requires a network of refueling (with gas) stations. 

So the critiques of Toyota are based on its lack of cooperation with other car makers in pushing for recharging networks to be made. In the short run, this makes marketing of electrical vehicles harder. In the long run, if Toyota should succeed with the fuel cell technology (far from a sure thing), it would undercut the entire electrical vehicle market. This is a kind of situation that often occurs when industry participants are locked in competitive battles and lobbying efforts together, and it is a situation that we still do not understand well enough through research. Nor are the firms good at handling it. The Daimler CEO was clearest in expressing what they thought Toyota should do about no-emission vehicles: "we have to set the stage for this industry change together." Good idea, except that Toyota remembers that car makers compete and that Toyota has won with new technology once before.

Boston, W. 2014, “Daimler, Renault Chiefs Knock Hydrogen Cars.” Wall Street Journal, Oct 3 2014.

Toyota fuel cell car prototype. The hydrogen tank (mounted under) is very similar to the LP gas tank used in Japanese taxis, so it is likely that such commercial use is the first market target.

Friday, September 12, 2014

Forgive or Forget? How people react to firm misbehavior

Last week I wrote about the surprising news that asset management unit Skandia had made a decision that annoyed its customers (Once Bitten Twice Shy? How Reputation Problems hit Mutual Fund Manager Skandia Again). The news are surprising because Skandia is familiar with customer reactions to reputation problems, as it is just over 10 years ago that it was hit by asset withdrawals as a result of a scandal in similar-name (but different management) firm Skandia Insurance. Part of what it learned is that customers leave and don't come back soon; we estimated that a typical fund would take three years to get back to the asset levels it had before the scandal. This would suggest caution, especially because the three years are after a scandal in which the mutual fund manager is actually innocent. It just happens to have the same name as its owner, an insurance firm with a scandal. Another mutual fund management company also owned by Skandia but differently named avoided this damage.

In addition to our research on how the scandal spread to many firms other than Skandia, Takako Fujiwara-Greve, Stefan Jonsson, and I also looked carefully at how customers withdrew and came back to Skandia during and after the scandal. The results will be published in International Economic Review, and are available here in longer form. We found that each mention of the scandal caused customers to withdraw money from the firm. So, either more scandal mentions meant that more people knew about it, or more scandal mentions meant that more people lost patience with Skandia. This is natural, but useful to show.

A more interesting part is the return of funds to Skandia. Here, we might assume that customers forgive – as a period goes without scandal mention, they will come back. Alternatively, we can assume that customers forget – with the meaning of forget being that new customers or customers of different firms than Skandia don’t really pay attention to a Skandia scandal, and may stumble into Skandia funds after a scandal. The idea of forgiveness appeals to us morally, and would be useful for a large firm that could have many potential forgiving customers after a scandal. The idea of forgetfulness is less appealing, and it means that large firms suffer more after a scandal than small firms would. So which is true?

First, it is clear that a model of forgiveness would have some relation between the people leaving and the people entering afterwards, typically in the form of some proportion of the people having left forgiving and returning. Forgetfulness is even easier to model because it just means that some portion of all customers available will start business with the firm, for example if they start disliking their current firm. And – our model of forgetfulness did in fact describe the return of customers very well. For Skandia, which was a large firm, forgetfulness was bad news because there would be more funds coming if customers were forgiving than if they were just forgetful. For other large firms it is potentially bad news too, because forgetfulness means that there is no loyalty or size advantage when the firm gets involved in a scandal. 

People may forgive other people, but firms they just forget.

Saturday, September 6, 2014

Once Bitten Twice Shy? How Reputation Problems Hit Mutual Fund Manager Skandia Again

Skandia is a an asset management unit owned by the insurance firm Old Mutual, and was in the press September 5 because it had decided to reverse a decision to move a sizable amount of funds (1 billion dollar) from its usual fund manager Invesco to a new company owned by former Invesco manager Neil Woodford. The reason for the reversal was protests by the owners of the funds it was planning to move, who did not think these funds could be just swapped across management companies without their agreement. Skandia’s temptation to move the funds makes some sense given that Mr. Woodford had star status and was thought to be responsible for much of the success of Invesco, the company he then let. The customers’ protests make sense because typically we normally dislike having our properties moved to different management without any say.

Is there any reason Skandia should have been a bit alert to the power of customers? Yes there is. Fund manager Skandia used to be a Swedish company, owned by a Swedish insurance firm also called Skandia. The takeover of Skandia by Old Mutual in 2005 followed soon after a period of customer protests involving significant withdrawals of funds had weakened the company. The reasons for these protests were all associated with a reputation for little attention to others. First, in 2002 a subsidiary firm was sold in a way that appeared to favor the mother company Skandia over other owners of the subsidiary. Second, top management pay received attention in the press because it was seen as unfair. Third, the press discovered that close relatives of the top management were given rental apartments at very low prices, a problem that was fully investigated during 2003. The press started referring to it as the Skandia scandal, and customers started withdrawing money from mutual funds.

In an article in Administrative Science Quarterly, Stefan Jonsson, Takako Fujiwara-Greve, and I have studied how the effect of this scandal spread, with particular attention to one important detail: the mutual fund management company Skandia did nothing wrong. All the three reputation problems were associated with the insurance firm that owned it. In spite of that, customers punished Skandia fund management. In fact, they even punished other fund management companies owned by (innocent) insurance companies, as well as fund management companies with owner companies that resembled Skandia in some way.  

The punishment for behaving in ways that customers disliked spread wide, to with no connection to the scandal. It lasted long, with the investment levels of Skandia taking three years to recover. Skandia is a good example of something we know already. Reputation effects on firms are powerful, and often overlooked. Overlooked so much, in fact, that even a firm that should have known better because it was once bitten, isn’t yet twice shy.  

Cox, Josie. 2014.Skandia U-Turns on Woodford Mandate. Wall Street Journal, 2014 Sept 5.

Sunday, August 10, 2014

The food tastes better if I can’t see her: Evaluation of female expertise

This is obvious to those who know Japanese food and obscure to others: Kaiseki is the fanciest Japanese food. No, it is not sushi or any other of the other straightforward and specialized kinds of food. Kaiseki is a course meal, with many courses, each of them having what we in western food would consider many courses. The first round of food is a bit like an appetizer, but in a kaiseki restaurant we would end up counting any number of small dishes on it. It is well worth trying out kaiseki if you have not already had it.

But I am getting too excited here and forgetting the story I was going to write. There is a kaiseki restaurant called n/naka in Los Angeles where the master chef stays out of view of customers. That is not so unusual in kaiseki restaurants, which often have chefs out of view, but there is a special reason: she is female, and some customers will be more satisfied if they can taste the food without knowing that it is made by woman. You see, kaiseki chefs are true experts, and nearly all male.

This sounds like a pretty specialized issue having to do with the norms of Japanese food (sushi chefs are also male, typically), but it is actually linked with what happens in work places as well, including important functions such as corporate research and development. We constantly evaluate the expertise of others, and in teams where expertise is required these evaluations are closely linked with work distribution and resulting effectiveness. A chef being assessed as less effective because she is female means fewer customers at the restaurant. An engineer being assessed as less effective because she is female could mean an inferior quality product – a problem for the firm, and also for you if the product happens to be the car you are driving.

So do we know when the evaluation is fair? This is a topic that there is much research on, and a typical finding is that it is harder for a woman to be evaluated fairly by others. Now, thanks to research on research teams by Aparna Joshi published in Administrative Science Quarterly, we know exactly how important the evaluator is in determining the fairness.  The results are actually quite simple. When a female assesses others, she will rate them higher the better their education is. That sounds simple and logical, and I bet you think you do the same. That could depend on your gender though: when a male assesses others, he will rate them higher when they are male and will ignore their education. That is a pretty big difference. These are research teams in a university, so of course we cannot know whether teams with less educated participants have a more educated way of assessing each other.

Actually, the story is a bit more complex because it depends on how strongly the evaluator identifies with his or her gender. Again the results are simple, but not really encouraging. A man who feels very manly will rate a woman below men regardless, and lower when she has more education. Yes, lower. A man who is more neutral will ignore her education and simply rate her lower than men regardless.  So, does this mean that firms should be careful about using women in roles that call for expertise to be correctly evaluated? Well, actually the opposite conclusion seems better. Women are actually good at evaluating women, and at evaluating men, so if teams had many women (especially in the supervisor role) they would function better. You could be better off driving a car designed by a team with mostly female engineers.

Fontoura, Maria. 2014. Meet Niki Nakayama, One of the World's Only Female Kaiseki Chefs. Wall Street Journal, Aug 8 2014.

Wednesday, July 2, 2014

Quirky Appliances: Network Advantage through Complementarity

Quirky is actually the name of a company. It collects ideas on innovative products, has "community members" on its list vote and comment on them, and arranges to have some of them manufactured and sold. The idea is roughly similar to the wisdom of crowds, with many people being smarter than a few, and has the same strength and weaknesses. The strength is that many people are in fact smarter than a few, especially in a topic that they have some knowledge about. The weakness is that a product that has never been made or sold is not a topic that most people are knowledgeable about. So, Quirky has some successes and some failures.

General Electric (GE) is also a company that has thrived on innovation and still does. It was an Edison company, after all, and it is in many industries where only the advanced and innovative stay ahead. But, GE is at least thought to be very traditional in its approach to innovation, with research and development staff making improvements to existing products and technologies, and doing some exploration of new ones. This means, of course, that a collaboration between GE and Quirky would mean two firms with the same idea of innovating, but totally different and potentially complementary approaches. Sound like a good idea? So did GE and Quirky, and they are working together.

So how is this working? Well, there is a new air conditioner called Aero that is pretty smart. It can do some things humans can't, in fact. The main point is that it is network connected and can be controlled by the smartphone of the user, which is effective but not truly smart. The truly smart options are little details like geo-fencing. Geo-fencing means that if you let it, the Aero will keep track of where your smartphone is, and use the location and movement to tell when you are headed home, so it can start up at the right time. Many think this is a neat idea, and the Aero has sold well.

But how did the collaboration work? In fact, GE stayed GE, if the information now reported is correct. They did collect some ideas from Quirky, but the rest of the design was thoroughly in-house. This is perhaps not surprising, because GE knows a lot about air conditioners. But it is also clear that GE benefited from the complementarity that gave these ideas. And GE is getting the idea that complementarity is not just something you pick up in a single alliance; more alliances make it better. In fact, many of the early sales of Aero were done through a collaboration with taxi company Uber. Selling air conditioners through a taxi firm? Truly complementary. Now GE is getting close to the type of hub-and-spoke network that I, Tim Rowley, and Andrew Shipilov discuss in our book "Network Advantage: How to Unlock Value from your Alliances andPartnerships."
Mann, Ted. 2014. GE's First Quirky Device Isn't Very Quirky. Wall Street Journal, July 2 2014.