An engineer says the deadline is impossible. A manager hears a bad attitude. An analyst finds a hole in the numbers. Someone questions their commitment. The problem enters the room as evidence and leaves as a personnel issue.
Imagine working in that organisation for long enough. You would learn which answers travel upwards. You would also learn which answers take their author down with them.
That is the company this essay calls the Songbun Corporation.
The name is deliberately uncomfortable. North Korea's Songbun system classifies people through family background and perceived political loyalty, with consequences for their rights and opportunities, as documented by the United Nations Commission of Inquiry. Corporate politics does not carry the same stakes as state persecution. The analogy concerns one mechanism: a hierarchy that lets perceived loyalty determine whose information deserves consideration.
The cost of being right
The distinction between useful criticism and disloyalty can become remarkably convenient for whoever receives the criticism.
In a field study and two experiments, Ethan Burris examined how managers responded to employee voice. Challenging voice received worse performance evaluations and less endorsement than supportive voice. Perceptions of loyalty and threat helped explain those responses.
This does not make every critic correct. It raises a harder question about evaluation: how much of a performance rating describes the employee's contribution, and how much describes the manager's comfort?
A warning can be badly delivered and technically sound. A reassuring presentation can be beautifully delivered and wrong. An organisation needs to tell those things apart. Otherwise, it risks promoting the person who makes failure sound manageable over the person trying to prevent it.
Silence has a business model
An open-door policy costs almost nothing to announce. Walking through the door with unwelcome information can cost rather more.
Detert and Burris studied 3,149 employees and 223 managers in a restaurant chain. Managerial openness was associated with improvement-oriented voice, with perceived psychological safety helping explain the relationship. What managers did mattered to whether employees spoke up.
The practical implication is uncomfortable. Before blaming employees for withholding ideas, examine what happens to the ideas they do bring. Who gets interrupted? Whose warning becomes a complaint about tone? Who must produce extraordinary evidence to challenge a decision that required almost none to make?
There is no universal law that speaking up destroys careers. Another large field study found favourable outcomes when employees and managers agreed that upward voice was high. The problem is the incentive structure around a particular kind of information: the kind that challenges somebody with power over its author.
When silence protects a career, silence is a rational response. Calling it disengagement avoids asking who made candour expensive.
A graph cannot read a mind
It is tempting to answer this problem with another ranking system. Count messages. Map connections. Find the indispensable bridge and the employee who barely appears.
That can reproduce the original mistake with more decimal places.
Network measures describe relationships within the data collected. Their meaning depends on what those relationships represent. In research across 38 work groups, centrality in advice networks was positively related to individual performance, while centrality in hindrance networks was negatively related. Being central was not a verdict in itself.
A communication graph cannot, by itself, establish courage, loyalty, disengagement or value. Few recorded connections might reflect a specialist role, missing data or work performed elsewhere. Many connections might reflect expertise, unnecessary approvals or a system that makes everybody ask the same person.
This is a research argument about how organisations handle evidence. It is not a claim that Ghost Observatory can classify employees by their motives. Replacing a manager's prejudice with an unsupported label would defeat the point.
Make the warning testable
The useful question is what happens after a concern is raised.
Record the claim. Identify the evidence that would support or contradict it. Assign someone to investigate. Keep the decision and its reasoning available when the outcome arrives. If the warning was wrong, establish why. If it was right, correct the decision and examine why it struggled to get heard.
That requires more discipline than praising candour in a company handbook. It makes management's response open to scrutiny too.
We want organisations to gain better knowledge of their processes. That ambition has a human dependency: someone must be allowed to say that the process is failing. Detailed records are of limited use when the accepted interpretation must flatter the hierarchy.
The Songbun Corporation has a simple weakness. It can become very good at identifying who agrees with management while losing the ability to identify when management is wrong.
References
- United Nations Human Rights Council (2014). Report of the detailed findings of the Commission of Inquiry on Human Rights in the Democratic People's Republic of Korea. A/HRC/25/CRP.1, paragraphs 271–278.
- Burris, E. R. (2012). The Risks and Rewards of Speaking Up: Managerial Responses to Employee Voice. Academy of Management Journal.
- Detert, J. R., and Burris, E. R. (2007). Leadership Behavior and Employee Voice: Is the Door Really Open?. Academy of Management Journal.
- Burris, E. R., Detert, J. R., and Romney, A. C. (2013). Speaking Up vs. Being Heard: The Disagreement Around and Outcomes of Employee Voice. Organization Science.
- Sparrowe, R. T., Liden, R. C., Wayne, S. J., and Kraimer, M. L. (2001). Social Networks and the Performance of Individuals and Groups. Academy of Management Journal.