[English] In what way should carmakers decide a recall?*

Although the practice of recalling automobiles has known a distinct increase[1] since the nineteen-nineties, largely owed to the fact that electronic faults linked to the multiplexing[2] of vehicles have risen, such a decision remained, for one company, a major strategic choice, not only in terms of cost, but also, and perhaps especially, in terms of the brand image.
Just like in the nineteen-sixties, carmakers’ decisions focused on the analysis of cost advantages by weighing up, on the one hand, the compensation, which would cover the demand from possible victims of disasters attributable to the identified failure, and on the other hand, the technical cost of the recall. Such decision factors are today a lot more difficult to implement. The cost of injuries and interest spent on victims have increased dramatically. For example, an American consumer of an American fast food chain restaurant received 2.9 million dollars in compensation for being burnt by a cup of coffee. Furthermore, the cost of their reputation by failing to recall a vehicle can today prove to be fatal.
The financial equilibrium of carmakers exposed to fierce competition and carmakers coping with consumers who are better and better informed may be challenged through a new crisis, as demonstrated in the recent Mitsubishi affair. The constructor accused of covering up the existence of serious damages, which were likely to cause some serious accidents, is today finding himself confronted with serious difficulties financially and commercially. So even though the practice of recalls is widely developed, the Japanese firm preferred to keep the detected technical failures of the vehicles a secret. What can then be the determinants of a firm’s decision regarding recall?

Variable practices from one constructor to another

The decisive factors of recalls vary from one carmaker to another. They depend largely on its competitive position and particularly on the image, which the make has next to the consumers. Within the competitive context, which is occurring in more and more arguments, which exclude the talk of price (competition through quality), the decision to recall vehicles constitutes of a strategy combining certain cost elements at once (those linked to recall), insurance principles vis-a-vis future risks (avoiding possible compensation in the case of a disaster) but also reputable investments likely to generate either the costs or the gains in terms of image. Before launching a recall campaign, the constructor should not only anticipate technical costs directly linked to the recall itself, but also the probable consequences in terms of the company’s image. If the decision not to recall allows a saving on the former, he is not exposing himself to any less of a possible controversy, which could jeopardize the financial viability of the company. The choice, which firms make, is nevertheless closely dependent on their competitive position. For certain “top of the range” carmakers, the security and reliability of vehicles constitute the core of their competitive advantage. A possible recall can be analysed by their consumers as a sign of guilty slackness. It would however be in the constructor’s interest to run this risk. On the one hand, the guarantee of the vehicle’s quality is part of the moral contract bound between the contractor and the consumer. If the latter was not assured, he would definitely turn away from the make the moment the vehicle is renewed. On the other hand, risking a concealment of a possible fault would be for the company’s image unacceptable. However for certain carmakers, the admission of a fault can represent the cost possibly gaining the upper-hand over its secondary effects, beneficial of course, but uncertain and closely dependant on the basic image of the constructor. A constructor punished for a poor reputation in terms of quality can, through a recall, confirm the bad image, which the consumers perceived of their vehicles.
Conversely, it is possible for certain types of carmakers to benefit from proceeding with such recalls. The recalls can then be interpreted as proof of the constructor’s efforts regarding quality. It becomes less and less risky for a less reputable constructor to announce vast recall campaigns, not to mention at a time when the multiple faults linked to the multiplexing of vehicles drives reputable carmakers to organize numerous recalls. In a paradoxical way, such a constructor can even heighten his reputation if the controversies are made at the same time as the reluctance of more reputable carmakers to correct the failures which affect their vehicles. Furthermore it is convenient to keep in mind, that this type of constructor is in competition with companies whose brand image is in every respect comparable. Thus all of this additional effort concerning quality will allow him to bring out a competitive advantage.
Ultimately, it seems as though the recall can equally have very diverse consequences regarding the reputation of the constructor according to the common beliefs of the consumers. The recall thus constitutes of a risky strategy strictly speaking for the constructor who is running it. At a first glance, an appealing alternative can be to proceed in carrying out the necessary modifications to the vehicles in top secret, by returning them at random to the dealerships for reasons of reparations or maintenance within the context of campaigns often referred to confidentially as “campaign updates” or “modernisation.” The danger involved in a secrecy policy is illustrated through the difficulties which are encountered today, Mitsubishi, Japan’s fourth constructor. Not only will all the defective vehicles not be modified (none were taken care of in a dealership), but the consumers can legitimately feel deceived by a constructor, in whom they had placed their confidence. What is worse, is the absence of a systematic recall will not enable the constructor to lose any of the responsibility in the case of a disaster linked to the defect in question.

An approach of the automobile recall through the game theory

Placed in a repetitive interactive situation with consumers likely or unlikely to renew the confidence they had previously placed in the company’s products, the constructor finds himself plunged into a strategic interdependent situation such as how to deal with the game theory. This theoretical context is much more adaptable as it allows one to realise the risks associated with each sensitive decision made by the constructor. The game bounding the constructor and the consumer together can, for example, be considered as a cooperative game. The first engages in the security and the quality of its products, the second remains loyal to the constructor as long as he keeps to his commitments. From then on, the constructor’s decisions can no longer be made independently from the reactions anticipated by the consumers.

It is possible to demonstrate some of the main principles by enabling an explanation of the principal factors of the decisions concerning recalls, through an approach in the form of games. A recall campaign can be broken down into two steps. While the first being that the constructor chooses whether to recall his vehicles or not, the second step is that his decision will depend on how he anticipates the consumer to react when faced with his decisions. If the constructor carries out the recall, two issues are possible. The consumer can take disciplinary action against the constructor. The acceptance of damages varies according to the vehicle which is concerned. The sanction taken by the consumer will not renew his confidence towards the constructor at the time of his next buy and therefore “proceeds into competition.” In addition to the direct cost of the recall, the constructor should accept a cost to his reputation which will be expressed in the long term by a drop in sales. The absence of a sanction signifies that the consumer renews all of his confidence in the constructor and will remain loyal to the make. This signifies that the only cost endured is that of the recall. It is even possible that the recall improves the image of the constructor. In this case, the cost of the recall is compensated by the gain regarding the image. If the constructor chooses not to recall, two new case figures are to be considered. If the consumer discovers that the constructor has voluntarily hidden the fault, it is likely that he will be heavily sanctioned, as demonstrated in the Mitsubishi case. Regarding image, the impact is just as catastrophic insofar that not only the reliability is questioned but also the honesty of the constructor. If the constructor is not exposed, the lack of a recall maximises its gains. The reputation acquired by the constructor can then come to the rescue in turning away suspicions.
Which decision will a constructor come to when faced with these choices? All will actually depend on how much the constructor anticipates the consumer’s reaction to each alternative choice, which is conditioned by the tolerance threshold of the consumer. It is then necessary for him to assign probabilities to each type of possible reaction in order to determine which should be his optimal approach. By doing so he is in a position to compare the hope of the gains associated with each of these possible decisions. The anticipation made by the constructor as to how the consumers perceive the quality of his vehicles, just like the existing asymmetrical information is a detriment to the consumers, would then be a determinant in order to explain the firm’s decision.

Edouard BARREIRO, Fréderic MARTY, Patrice REIS

* This contribution is an extract taken from an article written for Automobiles and Components, November 2000, Recall or not to recall ? The Mitsubishi affair.
[1] According to the l’automobile magazine review of November 2004, two out of three breakdowns are attributable to the electronics. Furthermore, between the years of 1999 and 2004 the increase in faults in electronic order rose by 70%.
[2] We call multiplexing, the capacity to transmit data coming from several pairs of equipment (transmitters and receivers) onto a single physical medium (called high speed way). The first interest of multiplexing for the automobile is the simplification of the electronic system through the reduction of the number of wire connections.
(http://www.supinfo-projects.com/fr/2003/multiplexage/1/)