When Organisations Undermine Leadership: Responsibility, Authority and Why Development Is Not Enough

Universities, schools and other large organisations spend a great deal of resources on leadership development. People engage with a range of CPD programmes, are given mentors, encouraged to reflect on their leadership style, are taught to communicate, delegate, strategise and manage conflict more skilfully. There is good reason for much of this, because leadership does require judgement, self-awareness and a range of skills, all of which can genuinely be developed. However, there is a danger in locating leadership too completely within the person who occupies the role, as though every difficulty encountered must therefore be a difficulty of character or competence rather than a difficulty of context. In reality, sometimes the most important problem is not that somebody needs to become a better leader. On occasion, it is the organisation itself which has made effective leadership challenging to exercise.

I was reminded by a recent article in Times Higher Education by Vicki Baker, who interviewed twenty-five current and former associate deans across universities in North America and Europe. She found a noticeable between those who experienced their roles as enabling and those who found them frustrating or exhausting. This difference did not seem to sit in individual capability; it was shaped, to a surprising degree, by the environment created by the senior academic above them. Those who felt they were flourishing described a context of clarity around what was expected of them; ready access to the information they needed; permission to disagree; real delegated authority and confidence that their line manager and those with oversight would stand behind them when it mattered (Baker, 2026). Conversely, those who felt themselves to be struggling described a situation in which problems moved downwards while power stayed put. Baker calls this responsibility without authority, and names it as one of the more corrosive features a role can have (Baker, 2026). I’m sure this observation resonates well beyond universities.

Responsibility is not authority

A formal leadership role brings responsibility, but responsibility and authority are not the same thing; however, we so easily run them together when we describe someone’s job. Someone may be held accountable for a programme while having little real control over its budget; they may be expected to manage a team while unable to decide staffing levels. They may carry responsibility for strategy while discovering, again and again, that the decisions that actually matter get made elsewhere. Under conditions like these, a job title can promise a great deal more agency than the job itself contains.

I recognise this from the inside. I have held senior academic roles where the title suggested a considerable degree of autonomy and responsibility, while in reality the authority and independence I was actually able to exercise were much more limited. Decisions I understood to sit within my remit were in fact shaped elsewhere, and matters I thought were still open for discussion had, in practice, already been settled. This was both disempowering and curiously wasteful of the resource and experience being paid for.

There was also a less tangible dimension in my experience of this. Formal position does not necessarily confer equal access to the relationships, networks and forms of recognition through which institutional authority is often ritualised and sustained. It took me time to recognise that what I was experiencing was not simply a question of my own reputation, confidence or capability, but a gap between the responsibility apparently attached to the role and the authority and recognition available to me in practice.

Research on distributed leadership in higher education has raised similar questions. Richard Bolden, Georgy Petrov and Jonathan Gosling, looking across UK universities, distinguished genuinely collaborative and emergent forms of leadership from more top-down processes in which responsibilities were dispersed without a corresponding dispersal of power (Bolden, Petrov and Gosling, 2009). More recent research suggests that the problem prevails. In an Australian university faculty, Lizier, Brooks and Bizo found that academic leaders within schools could experience what was described as distributed leadership principally as the delegation of duties. Their study highlights the importance of clarity, hierarchy and trust in determining whether distributed leadership actually becomes meaningful in practice (Lizier, Brooks and Bizo, 2024). Tony Bush makes a related point in the context of higher education more broadly, observing that responsibility without sufficient power over areas such as finance and staffing can leave academic middle leaders frustrated and functioning largely as conduits for decisions taken elsewhere (Bush, 2022).

A senior leader may believe they have empowered their colleagues because they have delegated a great deal of work. Yet delegating work is not the same as delegating power. Real authority requires some capacity to decide, to commit resources, to set priorities, and to trust that decisions made properly within the bounds of a role will actually be respected once made. Without that, the leader becomes something closer to a conduit: expectations flow down through them, and difficulties flow back up.

Responsibility without authority is not empowerment but exposure.

Leadership takes place within an organisation

There is a wider body of research behind this. Edgar Schein’s work on organisational culture showed that formal structures and declared values only tell us part of the story, and that organisations also carry deeper, often unspoken assumptions about authority, relationships and risk that shape behaviour far beyond what appears in policy documents (Schein and Schein, 2017).

More recent organisational psychology has placed increasing emphasis on the relationship between people and the conditions in which they work. Job Demands–Resources theory, for example, understands work engagement and exhaustion as arising not simply from individual characteristics but from the interaction between the demands people face and the resources available to them. Recent developments in the theory explicitly incorporate person–situation interactions and multilevel organisational processes (Bakker, Demerouti and Sanz-Vergel, 2023). Similarly, a major review of the now extensive literature on psychological safety shows how people’s willingness to speak, learn, question and take interpersonal risks is shaped by leadership, relationships and wider features of the working environment (Edmondson and Bransby, 2023).

Put together, these different strands of research point towards the same conclusion: what a person is capable of doing cannot be fully understood by examining the person alone. People exercise judgement within cultures, take risks within relationships and learn whether authority is real from what happens when they attempt to use it. Leadership takes place inside an ecology of information, trust, resources, relationships and power, much of which remains invisible on an organisational chart.

Leadership is partly conferred

There is a further practical implication. Formal authority does not become effective simply because it has been written into a job description or role title. Some of the capacity to lead is conferred, and continually reconferred, by the people and systems around us.

Authority is sustained when senior colleagues share information freely, are clear about boundaries, respect the power they have delegated, provide the resources a role requires, and are visibly willing to back the person in it. It weakens when decisions are repeatedly second-guessed, when information arrives too late to be useful, when people learn they can simply walk around the role-holder, or when institutional backing disappears at exactly the moment it is needed most. The role might stay unchanged on paper while its actual authority drains away. This is part of why leadership can feel so precarious in practice. The office may hold formal authority, but its effectiveness is relational because other people have to recognise it for it to mean anything.

Uhl-Bien’s influential account of relational leadership challenged the idea that leadership resides simply in individual traits or behaviours, treating it instead as something produced through social relationships and processes (Uhl-Bien, 2006). More recent reviews of relational leadership research similarly emphasise the importance of the relationships through which leadership is enacted, while also noting the need to understand those relationships within their wider organisational setting (Scandura and Meuser, 2022). The same person, with the same intelligence and experience and skill, can behave quite differently in two different organisations. In one, they are trusted with hard decisions, given the information they need and supported when those decisions provoke resistance. In another, they become hesitant, over-consultative, and increasingly anxious about acting beyond whatever has been explicitly signed off. It would be easy to call the second version a failure of confidence, and sometimes it will be, but sometimes the organisation has simply taught that person that independent judgement is not safe to exercise, and they have imbibed that message so well it has become part of the pattern of their behaviour.

Organisations can therefore become very good at addressing the behaviour of individual leaders while leaving untouched whatever it is about the system that keeps generating the same problem in new people. Sending the leader on another course can become a kind of organisational sleight of hand. We seek to improve the person but thereby avoid the harder question of whether something about the system itself needs to change.

The paradox of the indispensable senior leader

Baker’s article raises a further, more uncomfortable question. How should we actually judge senior leaders? We tend to judge them by what happened on their watch. Did performance improve? Were budgets balanced? Were strategies delivered? Did the organisation survive its crises? These are reasonable questions, but they can conceal something else worth asking, which is what happened to everyone else’s capacity to lead while this person was in charge.

Baker suggests senior leaders might also be judged by whether the people around and beneath them grew more capable of leadership during their time in post: whether they gained judgement, confidence and real authority, and whether they were trusted with steadily more consequential decisions (Baker, 2026).

This exposes a genuine paradox. A senior leader who becomes indispensable can look, on the surface, extraordinarily successful. Every decision seems to pass through them. They hold the institutional memory. People wait for their permission. Problems get solved because they step in. Their absence, when it comes, is felt everywhere.

Yet indispensability is not the same thing as good leadership. It can just as easily be a sign that authority and capability were never really developed anywhere else, that an organisation has become dependent on one person’s judgement precisely because nobody else was given a real chance to exercise their own.

The strongest senior leadership, then, might be judged partly by what becomes possible without that leader’s direct intervention. Building leadership capacity elsewhere means giving away some authority, and tolerating the discomfort of watching other people exercise it differently than you would have.

Where coaching fits

This matters for coaching.

There is now a substantial evidence base for workplace and executive coaching, although questions remain about precisely how, for whom and under what conditions it produces its effects. Earlier meta-analytic and systematic reviews found broadly positive effects across skills, affective outcomes and individual performance (Jones, Woods and Guillaume, 2016; Athanasopoulou and Dopson, 2018). More recently, de Haan and Nilsson’s meta-analysis restricted the evidence to randomised controlled trials. Across 37 studies, 39 independent samples and 2,528 participants, they found a statistically significant moderate overall effect for workplace and executive coaching across leadership and personal outcomes (de Haan and Nilsson, 2023). The findings were not unqualified. Effects were greater for self-reported than externally observed outcomes, and the authors found evidence of publication bias. Nevertheless, the study provides comparatively rigorous evidence that coaching can produce meaningful effects.

Writing a decade after their original meta-analysis, Woods, Jones and Guillaume (2026) similarly reflect on the development of the coaching evidence base and the questions that remain for research and practice. The issue is therefore no longer simply whether coaching can have beneficial effects, but how those effects arise, for whom, and within what organisational conditions.

So, coaching can make a difference. The real question is what kind of difference we should expect it to make.

Coaching can be especially valuable in helping a leader work out what belongs to them and what belongs to the system around them. A leader will often present a difficulty as a personal one: they lack confidence, they are reluctant to decide, they find a particular relationship hard, they feel overwhelmed by responsibility or unsure of their own capacity to influence events. Sometimes that is exactly right, and there are assumptions or habits worth examining. But sometimes that framing quietly conceals an organisational problem dressed up as a personal one.

Coaching allows more discriminating questions to be asked. Do you actually have the authority to match the responsibility you carry? Do you have the information you need to lead? Are the boundaries of your role clear, or are they only assumed by your job title? Are you being unnecessarily cautious, or have you simply learned, from experience, that acting independently carries real risk? Are you avoiding a conversation that is genuinely yours to have, or being asked to fix something over which you have almost no meaningful control? If this is not yours to solve, what is the cost of continuing to behave as though it were?

The point is not to lift responsibility from the leader’s shoulders but to locate it accurately, so that what follows is aimed at the right target. From there, coaching can help someone distinguish what they might change in their own practice, what needs to be negotiated with others, and what might need to be challenged at an organisational level. It can help them make fuller use of whatever authority they do have, prepare for difficult conversations about role and resource, notice where they have been quietly accommodating a pattern that does not actually work, and decide what boundaries need to be drawn. Occasionally it can also help someone see that a role which is structurally incoherent will not become viable no matter how hard they try to inhabit it.

Anthony Grant’s research into coaching during organisational change is useful here. In a study of managers going through major change, coaching was linked to greater goal attainment, more solution-focused thinking, stronger change readiness, higher leadership self-efficacy and greater resilience (Grant, 2014). Grant also draws attention to the importance of focusing coaching upon goals and concerns that sit within the client’s sphere of influence.

That qualification is really the heart of the argument. There is a real difference between someone needing to get better at exercising the authority they already have, and someone having been handed responsibility for something they were never given adequate authority over. Both can produce frustration and self-doubt and poor outcomes that look, from the outside, remarkably similar. But they are not the same developmental problem, and treating them as though they were will waste a great deal of good coaching.

In the first case, coaching can help someone change how they act. In the second, it can help them see their situation clearly, work out what needs negotiating or challenging, and decide how they want to respond to circumstances that they cannot, on their own, repair. Coaching does not mend a broken organisational structure. What it can do is help a leader understand more clearly how they sit within one, and what, realistically, is theirs to change.

Coaching and organisational learning

There is one more thing worth saying. Coaching is usually thought of as something that happens strictly between a coach and an individual client, but what surfaces in the room often says something about the organisation as well as the person.

Recurring difficulties around delegation, authority, anxiety or conflict are not always simply a collection of individual problems. They may be manifestations of organisational culture. Schein’s work remains useful here, while contemporary research on psychological safety strengthens the point: what appears to be an individual reluctance to act may reflect deeper assumptions about authority and risk, or a working environment in which questioning, disagreement and interpersonal risk have become costly (Schein and Schein, 2017; Edmondson and Bransby, 2023).

This means leadership coaching sometimes sits right on the boundary between helping one person and noticing something about the whole organisation. That needs care. A confidential coaching relationship must never turn into an informal intelligence operation for the employer, and no coach should presume to diagnose an entire institution from one person’s account of it. But coaching can still help a leader recognise a pattern and decide, on their own terms, whether it is something worth raising more widely. The influence can run both ways: the organisation shapes the leader who walks into the room, and what that leader comes to understand may, afterwards, let them act differently on the organisation itself.

Development is not a substitute for organisational responsibility

None of this diminishes the value of coaching; rather, it clarifies its proper place. Coaching can strengthen judgement, self-awareness and agency. It can help someone make fuller use of the authority they possess, recognise assumptions that are constraining them, prepare for difficult conversations, and see where they may have accommodated themselves too readily to an organisational pattern. What it cannot responsibly do is turn a lack of authority into a problem of personal resilience, or make a structurally incoherent role coherent by changing the person who occupies it.

This distinction matters for organisations commissioning coaching. When a leader is struggling, the question should not simply be what development they need, but what kind of problem is actually being presented. It may be a developmental need, a difficult relationship, a lack of confidence, a problem of role design, or a mismatch between responsibility and authority. Several of these may be present at the same time. Sending an individual for coaching before asking such questions risks making the intervention itself part of the process by which an organisational problem is individualised.

There is therefore no contradiction between investing seriously in coaching and recognising the limits of what coaching can achieve. Indeed, good coaching should make that distinction clearer. Its purpose is not simply to make people better able to tolerate whatever circumstances an organisation creates. It can help leaders understand the circumstances they inhabit, recognise where they have agency, decide where they need to act differently, and identify what may need to be negotiated or challenged.

Perhaps leadership development is best understood, then, not as work upon a person in isolation, but as work at the intersection of person, role and organisation. Coaching can contribute to that work because it provides somewhere in which all three can be examined together. The organisation nevertheless remains responsible for the conditions within which leadership is exercised.

A capable leader still needs authority that is real, information that arrives in time, sufficient room for judgement, relationships in which disagreement is possible, and confidence that legitimate decisions will be supported. Development may make someone better able to recognise and use those conditions, but it cannot indefinitely compensate for their absence.

The question, then, is not simply whether we are developing better leaders. It is whether we are creating organisations in which the people we develop are actually able to lead.

REFERENCES

Athanasopoulou, A. and Dopson, S. (2018) ‘A systematic review of executive coaching outcomes: Is it the journey or the destination that matters the most?’, The Leadership Quarterly, 29(1), pp. 70–88. doi: 10.1016/j.leaqua.2017.11.004.

Baker, V.L. (2026) ‘Want better university leaders? Stop focusing on the leaders’, Times Higher Education, 14 September 2026. Times Higher Education (THE)

Bakker, A.B., Demerouti, E. and Sanz-Vergel, A. (2023) ‘Job Demands–Resources Theory: Ten years later’, Annual Review of Organizational Psychology and Organizational Behavior, 10, pp. 25–53. doi: 10.1146/annurev-orgpsych-120920-053933.

Bolden, R., Petrov, G. and Gosling, J. (2009) ‘Distributed leadership in higher education: rhetoric and reality’, Educational Management Administration & Leadership, 37(2), pp. 257–277. doi: 10.1177/1741143208100301.

Bush, T. (2022) ‘Middle leadership in higher education: Challenges and opportunities’, Educational Management Administration & Leadership, 50(5), pp. 737–738. doi: 10.1177/17411432221112489.

de Haan, E. and Nilsson, V.O. (2023) ‘What Can We Know about the Effectiveness of Coaching? A Meta-Analysis Based Only on Randomized Controlled Trials’, Academy of Management Learning & Education, 22(4), pp. 641–661. doi: 10.5465/amle.2022.0107.

Edmondson, A.C. and Bransby, D.P. (2023) ‘Psychological Safety Comes of Age: Observed themes in an established literature’, Annual Review of Organizational Psychology and Organizational Behavior, 10, pp. 55–78. doi: 10.1146/annurev-orgpsych-120920-055217.

Grant, A.M. (2014) ‘The efficacy of executive coaching in times of organisational change’, Journal of Change Management, 14(2), pp. 258–280. doi: 10.1080/14697017.2013.805159.

Jones, R.J., Woods, S.A. and Guillaume, Y.R.F. (2016) ‘The effectiveness of workplace coaching: a meta-analysis of learning and performance outcomes from coaching’, Journal of Occupational and Organizational Psychology, 89(2), pp. 249–277. doi: 10.1111/joop.12119.

Lizier, A., Brooks, F. and Bizo, L. (2024) ‘Importance of clarity, hierarchy, and trust in implementing distributed leadership in higher education’, Educational Management Administration & Leadership, 52(4), pp. 901–915. doi: 10.1177/17411432221105154.

Scandura, T.A. and Meuser, J.D. (2022) ‘Relational dynamics of leadership: Problems and prospects’, Annual Review of Organizational Psychology and Organizational Behavior, 9, pp. 309–337. doi: 10.1146/annurev-orgpsych-012420-091249.

Schein, E.H. and Schein, P.A. (2017) Organizational Culture and Leadership. 5th edn. Hoboken, NJ: Wiley.

Uhl-Bien, M. (2006) ‘Relational Leadership Theory: Exploring the social processes of leadership and organizing’, The Leadership Quarterly, 17(6), pp. 654–676. doi: 10.1016/j.leaqua.2006.10.007.

Woods, S.A., Jones, R.J. and Guillaume, Y.R.F. (2026) ‘Coaching and its effectiveness in organizations: Reflecting on a decade of research, looking forward to future challenges’, Journal of Occupational and Organizational Psychology, 99(2), e70115. doi: 10.1111/joop.70115.