Notes on Trust
and why it matters in corporations
Brigitte Jordan, Ph.D.
Consulting Principal Scientist
Palo Alto Research Center
Last edits: Wednesday, September 11, 2002
I have long been mystified by the clarion call for “trust” that issues from corporations when something is not quite right in their organization. The remedy is often a “culture czar” or a team of “culture brokers” who are charged with designing “a culture of trust” into the next change strategy -- with calls to action that often don’t understand what kind of a thing trust actually is.
So what is trust? We could say that a person or organization trusts another when both sides have reason to expect that neither will take advantage of the other, and, whenever possible, will even do things that advance the other’s interests.
Thus, when I trust my doctor, I expect that she will give me the best advice she is capable of and that she will not order a procedure that is unnecessary or damaging, even if that would be to her financial advantage. When I trust my babysitter, I believe that he will do everything humanly possible to protect my children from harm. When I trust my employer, I expect that he will not reorg me out of existence. However, as all of us know only too well, trust may be misplaced. There is always the possibility that expectations will be disappointed, even in the closest relationships.
Where do we find trust?
Trust may exist between groups of people, between individuals and groups or between particular individuals. Trust between individuals is found in highest concentration in close-knit social units such as the nuclear family, a swat team engaged in a crucial mission, a sports team when it clicks, or a community of practice passionately engaged in some kind of skunk works. Thus trust relations would be expected to be more likely within a particular sociocultural group than between such groups.
Since trust is calibrated to the potential risk if the other party does not act as expected, it has a probability distribution and a density. There is a texture to it. I may expect that you will give me unbiased feedback on this paper but I may not trust you to risk your life for me. Unconditional trust exists only in the most intimate relationships if it is achieved at all.
In situations where there is a power differential, that is, when one party is vulnerable to the actions of the other, there is a risk of harm. In such situations, trusting the other simplifies the range of possible actions by eliminating resistance and other actions that might be defensive or obstructive.
How do we build trust? Contrary to corporate organizational experts it can’t be “designed.” It has to be built -- little by little. History matters here. A lot. In a dynamic, non-routine work environment trust is essential. People must feel they can take risks and that their vulnerability will not be taken advantage of. “We’ve had bad experiences with the engineering department” is an indicator of distrust that will require major effort to dispel. Where there is a history of bad relations between labor and management, distrust tends to shape interaction. When there has been massive downsizing, workers will regard company assurances with suspicion and skepticism.
Trust is generated in the course of an extended history. It produces repeatedly told “eye-witness” stories about what the other party is up to, providing evidence that the interests of the two parties in some way complement each other. Trust between ethnic groups and organizational subunits usually depends on a long-standing shared tradition of mutually beneficial interactions. Trust between consumers and a company (a “brand”) is always based on an extended set of experiences with the company’s products, sometimes going back generations. (“I’m a third generation Maytag user”). Though not recognized as such, trust relationships enter into everyday conversations and to a considerable extent are built in and through those conversations. In that way, the experience of others (in the stories about that CEO, that department, that product) takes on a social life. It becomes available to an entire community who may lack personal experience with the trusted entity. The lunchtime story about the Xerox rep who stayed till midnight to get a broken-down printer going again reinforced her listeners’ trust in the company. (“They do what’s right.” “You can depend on them.”)
And then there is distrust. Distrust is worse than just the absence of trust. It rears its ugly head when you actively expect that the other guy will take advantage of you, when you have reason to suspect that, given a chance, your “friends” will further their own interests and ignore yours.
Distrust generates an expectation of potential harm. Though it may sometimes be based on prejudice or stereotype, it often has its roots in memories of past negative interactions. While trust is rooted in positive expectations about the other party’s motivations and behavior, distrust is generated when one party has reason to suspect that the other will act selfishly. Such expectations are often grounded in a history of past interactions that do not allow projecting a mutually beneficial course for the future This is always the case when implicit or explicit promises have been broken.
Distrust arises when there is lack of predictability. We don’t trust people (or organizations) when they act erratically, in ways that are unpredictable for us. That’s why we don’t trust strangers. Years ago the sociologist Harold Garfinkel devised a set of brilliant experiments to investigate what happens when implicit trust is broken. He sent his students out to undermine people’s expectations about how others in their environment would act. Students would, for example, act like boarders in their own home or treat a shopping customer in a store like she was a sales person. Now, how would you feel if your wife all of a sudden asked for permission to put her yogurt in the refrigerator? Or a fellow customer asked you (at first politely, but then adamantly) that you show her a dress in a different size? Garfinkel’s breaching experiments seriously undermined the victim’s implicit expectations about how others would (and should!) act. The massive incongruity generated by these interventions generated not only grave distrust in the victim but also considerable social distress and acute anxiety in the perpetrators (Garfinkel 1963:189).
It is interesting that both trust and distrust are mechanisms of social control. They both reduce social complexity. Both point a way to action and thereby reduce anxiety in uncertain situations. Distrust, however, has serious drawbacks. It consumes an inordinate amount of energy, produces competitive behavior and leads to a “cover-your-ass” attitude. As we will see, this has important consequences
Generating organizational distrust
That broken promises lead to distrust is clear. But in corporations distrust is often created unknowingly when organizational boundaries are breached. All organizational groupings have established work practices that govern information flow across their boundaries. The control of this information is an important part of the social relations between work groups in organizations. Organizational change can disrupt or destroy the boundary maintenance mechanisms that are important for managing intergroup social relationships since it threatens to remove control of information from the local group and thereby their ability to manage their boundaries. The increased risk represented by loss of control over boundaries can fly in the face of management efforts to encourage cooperation across work groups. When there is distrust of what is beyond the local domain, people become anxious about border control. They also become defensive and risk- averse, preferring to stick to the true and tried. This obviously inhibits the sharing of information across boundaries, which is a prerequisite for effective coordination and collaboration.
Distrust also may come from a misapplication of Quality principles. Deming and his followers understood Quality as a way to give workers and work groups the resources for improving their own performance. But many corporations are using quantitative performance measurements to evaluate and rank performance competitively. There are strong indications that such use leads to conditions of distrust, not only between frontline workers and management but also between co-workers and between teams, since they are thrust into a competitive rather than a learning situation.
Deming’s brilliant idea was that the workgroup would decide which of the hundreds of possible measurements of their work process were most valuable to them for feedback about their performance. What happened, however, in many companies is that managers began to use these figures for their own purposes. Baker (1994) reports that at Ford, when management asked their plants for more performance figures, new indexes were developed and measurements were compared across plants. A whole new crew of bureaucratic intermediaries arose who computed averages and compared hi’s and low’s across plants and across quarters. This in turn led to new levels of accountability – no longer were the data used specifically to improve the workgroup’s performance in their work processes, but plant managers now had to explain why their figures weren’t as good as those of the next plant and what plans they had made to improve them. The game in each plant changed from making improvements to making the plant look good against its neighbors (Baker 1994).
A similar story is told by Jody Gitell who analyzed what happened when American Airlines instituted a detailed performance measurement system. Stringent performance standards for everything from baggage handling to customer complaints were reviewed on a daily basis. Failure to meet these standards was perceived to result in punitive action. Every day, numerical performance information was sent from the field to headquarters, but there was little discussion generated by these numbers and consequently little learning took place. The unintended effect was a “culture of fear” where employees looked out for themselves in order to avoid recrimination rather than focusing on the shared goals of on-time performance, accurate baggage handling and satisfied customers (Gittell 2000).
Gitell and other organizational theorists have argued that quantitative performance measurements inevitably generate dysfunctional behavior and low levels of trust. Indeed a system of individualized accountability is virtually guaranteed to produce low trust by setting up conflicts over who is responsible for problems and deficiencies.
It is no surprise that when performance data get used to punish and reward, there is a very strong temptation to manipulate the numbers. This is something we have seen in every workplace study that we carried out at PARC and IRL.
· In an airlines operations room, error codes are assigned with an eye to a whiteboard on which errors so far are plotted against the monthly allowable number of errors in each category.
· Telephone repair people use a “customer-not-at-home” code for justifying service calls they aborted under the pressure of an overwhelming workload.
· Call center operators forward a troublesome call to cut it short so they can make their targets.
· Teachers “teach to the test.”
In the competition for looking good, the game becomes one of making the numbers look good rather than improving the learning or work process (Jordan and Putz 2002)..
Why should corporations care?
One reason corporations need to worry about trust is that trust is essential for smooth collaboration internally (between co-workers, teams, communities of practice and hierarchical levels) and externally (with suppliers, customers, and regulatory agencies). In situations of distrust, workgroups may deliberately send out false information to strengthen their position (“ doctoring the numbers”, “gilding the lily”) because they are afraid of being taken advantage of. At the same time, the recipients of this information may discount what they do get and attempt to get quality information in other ways.
In one company we studied, the executive team knew very well that the information they had in their customer data base was fundamentally flawed. Sales people submitted insufficient information and didn’t update changing accounts in order to protect themselves from having their territory reorganized. The executives, fully aware of that fact, hired a consulting firm to clean up the data base, a strategy that led to a prolonged period of guerilla warfare between the sales people and the hapless consultants in search of those elusive “non-virtual clients”.
Obviously, under conditions of distrust, much energy gets expended on patrolling boundaries. Much of this energy is focused on building justifications for the group’s current behavior. The outcome is a great reluctance to move into uncharted solution spaces with the other organization, an insistence on the old ways, on “we know this works”, etc. In situations rife with distrust, there is no exploratory behavior, no experimentation, no risk taking, no out-of-the-box thinking.
Building trust in corporations
It should be clear by now that trust requires a shared system of values and shared assumptions about how people will behave. Fostering close-knit groups at all levels of the company is one obvious way in which intra-company trust can be created and nurtured. Often such groups already exist though they do not appear on the organization chart. It might actually require an ethnographer to ferret them out.
Another, though more costly way to trust is through elaborate legal arrangements. We trust our government for certain things (but not others) because of constitutional guarantees of various sorts and because of potential recourse to an elaborate legal system. In companies, trust between hierarchical levels is often supported by quasi-legal arrangements about job security, freedom from discrimination, ethical standards and the like. This apparatus, however, is most suitable for guarantees and assurances concerning the high-level structure of a relationship. It becomes unwieldy and very quickly removed from the realities of everyday interactions if it reaches too far down in the organization, as shown by the examples of Ford and AA cited earlier.
What then are some practical strategies for building and nurturing trust in companies?
1. Become aware of the informal groupings in your company, the communities of practice, the social networks within which trust is high and take care not to inadvertently destroy them. Nurture existing relationships of trust.
2. Make information freely available within the company. Support information exchange through lateral (peer) channels. Beware of simply cascading official news and programs through the company or disseminating information on a controlled, as-needed basis.
3. Discourage competition which tends to generate distrust of co-workers and parallel groups, but encourage collaborative performance.
4. Understand the boundary maintaining mechanisms between the parties.
5. Before orchestrating a culture change, such as a new business venture, merger or reorganization, try to ascertain the social connections between the parties that are expected to become involved with each other in the changed environment and assess the level of trust and distrust.
6. Build a history that counteracts the distrust-justifying stories in circulation. This takes time. Assess to what extent the stories reflect conditions that persist in the company. Be honest about mistakes, remedy them and generate counter stories.
I know I’ve just complicated the matter way beyond “designing a culture of trust” by hiring a trust officer. You may need, in fact, an ethnographer who can help you assess the invisible subterranean relationships in your organization, those undocumented networks that support effective work as well as undermine it. In the new economy where information is no longer easily contained within the old boundaries, understanding how to nurture existing relationships of trust and how to provide opportunities for building new ones is absolutely essential.
Some writings that helped my thinking here:
1963 Dangerous Liaisons: Trust, Distrust, and Information Technology in American Work Organizations. Human Organization 58:3:331-346.
1963 Springing Ourselves from the Measurement Trap. Pp. 454-457 in Peter Senge et al. The Fifth Discipline Fieldbook. New York: Doubleday.
1963 Conception of, and Experiments with, “Trust” as a Condition of Stable Concerted Action. Chapter 7, pp. 187-238 in: Motivation and Social Interaction, O.J. Harvey, ed. New York: Rowald Press.
Gittel, Jody Hofer
2000 Paradox of Coordination and Control. California Management Review 42:3:101-117.
Jordan, Brigitte and Peter Putz
2002 Rethinking Assessment as Practice: Notes on Measures, Tests and Targets. Unpublished ms.