Does the notion of ‘trust’ have any place in alliance metrics?
It is striking how frequently the analysis of the health of alliances rests substantially on the notion of ‘trust’. According to this analytical approach: if there is a high degree of trust between the alliance partners then the alliance is, somehow, in better shape and more likely to stay on track than an alliance with a lower degree of trust between the parties. However, this approach throws up a number of problems.
What is trust? Doz and Hamel illustrate the definitional problems involved in the use of the term ‘trust’;
‘Trust’ means different things to different people. To some it means putting one’s interests in the hands of a partner, confident that the partner will act as a good steward. In business alliances, trust may take the form of giving another company the use of one’s best resources, trusting it to use them in the best interests of the alliance…Another form of trust rests on confidence in one’s ability to predict the behaviour of another party. This confidence, in turn, is based on a superior understanding of the other party’s situation and self-interest. To still others, trust is a matter of ethical standards.
Trust is a simple, powerful, and emotionally laden words. As such it can be a source of confusion if used prescriptively for alliances. To avoid possible confusion, we suggest that partners replace trust with ‘enlightened self- and mutual interest,’ a concept that drove British diplomacy and foreign policy in the nineteenth century. As a basis for alliances, this is more appropriate than the idealistic but perhaps hollow concept of trust.*
Sage advice.
What are we measuring? Trust can reside in individuals or organisations. Typically, any level of trust between alliance partners will invariably involve some element of both the individual and the organisational. Unfortunately, however, analysis of the level of trust in an alliance frequently rolls up both trust in the individuals and trust in the organisation into one aggregated measure of the overall trust between the organisations. Nonetheless, the difference is important. Where trust has broken down between individuals those individuals can be trained, counseled or replaced. However, it is not at all clear what could be done to build trust if the distrust is directed at the organisational level; if, for example, the team on one side trusts the individuals on the other side but does not trust their organisation to support its employees.
How do we measure trust? The most direct way of measuring trust in an individual or an organisation is simply to ask ‘Do you trust X?’ The question can be extended to make it clear what our definition of trust is, for example, ethical standards or follow through on commitments.
However, some analytical approaches to measuring trust eschew the direct approach, preferring rather to create a synthetic value for the level of trust. Those adopting this approach typically create the synthetic value by applying an algorithm to the responses to a number of questions. The devil-in-the-detail in this approach, of course, is in the algorithm itself. So before you base your analysis of your partner feedback on a synthetic value of trust based on an algorithmic aggregation of the feedback from a number of questions make sure that you ask yourself the following questions: How do I know that the most appropriate questions have been used for the inputs and that no appropriate questions have been omitted?, How do I know that the weighting given to each question or each respondent are justified?, and How, flat out, do I know that the algorithm is some meaningful way right?
Does the level of trust have any predictive value or, to put it another way, is it more likely that a collaboration with a lower aggregate level of trust is more likely to fail than one with a higher aggregate level of trust? As with so much in this field the answer is that it probably depends on a number of factors. Clearly one such factor is how important trust is to the success of the collaboration. Research-driven collaborations, for example, probably require a higher degree of trust than distribution alliances. Another factor is the nature and the extent of the breakdown in trust. In a research-driven collaboration it is a disaster if the head of research in each organisation distrusts each other. On the other hand, the head of finance could probably exhibit a high degree of distrust with little damage to the prospects for the alliance.
* Alliance Advantage, Yves L Doz and Garry Hamel (1998) page 28.
