03 Nov 2026 Corporate Culture

The Leadership Anxiety Tax: How a Fear of Surprises Becomes a Culture of Control

By Andrew Baker · Group CIO, Capitec Bank · 3 November 2026

1. “Never Surprise Me”

Many years ago, a few weeks after I’d joined a new organisation, a senior leader pulled me aside and told me, “Never surprise me.” It wasn’t advice, and it certainly wasn’t a friendly tip from someone looking out for the new guy. It was an instruction, delivered with just enough menace that I understood there would be consequences if I ignored it. What he was really telling me was that I should never embarrass him, never let anything reach him that he hadn’t already seen coming, and never do anything that might leave him looking anything other than firmly in charge. My job, as he saw it, was to keep him in the power seat.

I remember walking away from that conversation slightly unsettled and genuinely puzzled, because I couldn’t work out why a surprise should automatically be a bad thing. What if my team stumbled onto something extraordinary, or found a much better way of doing something, or cracked a problem that everyone else had quietly filed under impossible? I would have thought those were exactly the kinds of surprises any leader would want to hear about. As I got to know him better, though, I came to realise that his attitude to surprises told you most of what you needed to know about the way he led. He was one of the most difficult people I’ve ever worked for, someone who needed not just to be in control but to be seen to be in control, and at the time I had no real sense of how much that one sentence revealed about the environment he would go on to build around himself.

The conversation came back to me recently while I was watching a documentary, when one of the contributors made an observation that I haven’t been able to stop thinking about. Paraphrasing slightly, the point was that one of the first things to disappear under tyranny is our capacity to be surprised. Sitting there, I realised that the instruction I’d been given all those years ago had never really been about surprises at all; it had been about control. And that led me to a question that is much more ordinary, and much more relevant to most of us, than anything involving tyrants: what happens when a perfectly normal amount of leadership anxiety slowly becomes the way an organisation operates?

2. What “No Surprises” Should Actually Mean

I want to be fair to the phrase before going any further, because “no surprises” has a perfectly respectable meaning and I suspect most experienced leaders, myself included, have used it at some point. In its healthy form it simply means that if you know about a material risk, a serious incident or some genuinely bad news, you don’t sit on it, and you certainly don’t let me find out about it from a regulator, a customer or a journalist. That’s a reasonable thing to ask of anyone, and in a bank it isn’t even negotiable, because bad news that travels slowly is about the most expensive thing an organisation can carry.

The trouble is that the phrase almost never stays in that narrow lane. “Never hide a risk from me” and “never let anything happen that I didn’t see coming” sound similar, but they produce completely different organisations. The first is a rule about candour, and it tends to make places safer, because it rewards people for putting their hand up early. The second is a rule about predictability, and it tends to make places more fragile, because the only reliable way to make sure a leader is never surprised is to make sure nothing happens without them hearing about it first. It’s the second version that worries me, and it’s the one the rest of this post is about.

3. The Leadership Anxiety Tax

Leading anything of real size means living with uncertainty, because you’re accountable for outcomes you can’t personally control and most of what matters happens while you’re not in the room. Some leaders find that feeling genuinely hard to bear, and rather than learning to sit with it they build machinery to make it go away. It starts innocently with a report, then a dashboard to go with the report, a steering committee to review the dashboard, a weekly update for the people who missed the committee, and eventually a spreadsheet that carefully explains what the previous spreadsheet already explained. Each request looks sensible on its own, so nobody pushes back, and over a year or two the organisation quietly rearranges itself around the emotional needs of the people running it.

I think of this as the leadership anxiety tax: the time, attention and creativity that teams hand over in order to reassure their leaders that everything is under control. What makes it so persistent is that the people levying it rarely see it. They believe they’re strengthening governance and staying properly informed, and sometimes they are, but quite often their own discomfort is simply being converted into running costs that get paid by everybody below them.

It’s worth putting a rough number on that, purely as an illustration. Imagine a hundred managers who each spend three hours a week preparing updates whose main purpose is reassurance rather than decision making. Over a year that’s 15,600 hours, which is roughly the working year of eight or nine full time people, and that’s before you count the time spent reading those updates, sitting in the meetings that discuss them and chasing the ones that arrive late. The exact figure in any real organisation will be different, but the point stands: anxiety at the top has a measurable price, and it’s paid in the capacity of everyone else.

4. The Reports Nobody Reads

I’ve often wondered how many reports in large organisations exist for no better reason than that somebody once asked for them. A weekly update gets created after an awkward meeting or a dashboard appears after an incident, and six months later nobody can remember why it exists, yet somebody is still producing it every Friday and somebody else is still chasing them when it’s late. By then it has stopped being a tool and become a ritual, and rituals like these give an organisation the feeling of communicating while crowding out the real thing. People polish updates when they could be working through the problem together, and they learn that a reassuring green indicator causes far less grief than an honest explanation of something complicated.

The test I’d encourage every leader to apply is simple: when did this report last change a decision I made? A report that informs a decision is worth having, but a report that exists mainly to soothe somebody’s nerves is waste, however professionally it’s formatted, and if it hasn’t changed anything for months it’s only fair to ask why people are still spending their evenings putting it together.

5. Visibility Is Not the Same as Control

None of this means leaders shouldn’t want visibility, and in organisations that run critical services there are serious obligations around risk, resilience and compliance that nobody sensible would argue against. But visibility and control are different needs that produce very different behaviour. A leader who wants visibility asks for a handful of measures that genuinely matter, clear thresholds for escalation and a regular chance to talk properly about the big issues, whereas a leader who needs control wants to know what everyone is doing, why, when it will be finished and why there hasn’t been an update since yesterday afternoon. The first creates a place where people can get on with their work, and the second creates a place where people spend a surprising proportion of their week explaining that they are, in fact, working.

6. The Importance of Tolerating Lag

If I could persuade leaders to take just one idea away from this post, it would be this one, because I think we badly underestimate how much value there is in letting information mature before we demand an update on it. When a team runs into a genuinely hard problem, they almost never understand it straight away. What they have at first is a collection of symptoms, a few half formed hypotheses, some incomplete evidence and usually two or three competing explanations, and they need time to dig in, argue about it, test their assumptions and gradually work their way towards an account of what’s actually happening.

An anxious leader tends to want the update before that process has run its course, so the team gives them one, and because they don’t yet know the answer, what they give is their best guess at that moment. The guess then gets repeated upwards, forwarded, challenged and turned into yet another reporting obligation, and before long the team is spending its energy defending a hypothesis that it really ought to still be testing. I’ve seen this most vividly during technology incidents, where the first explanation of an outage is almost never the complete one. Something failed, a deployment went out, a dependency fell over or a database was overloaded, but those are observations rather than causes, and getting to the real underlying cause takes proper investigation. If we push teams into certainty too early, we end up teaching them to confuse symptoms with causes, and we very often end up fixing the wrong thing with great confidence.

The distinction I’ve found most useful here is between two quite different kinds of lag. Escalation lag is the delay between someone knowing about a material risk and the right people hearing about it, and that should be as close to zero as you can possibly get it. Understanding lag is the delay between a problem showing up and the team genuinely knowing why it happened, and that isn’t a failure of any kind; it’s simply how long good work takes. The best leaders I’ve worked with are ruthless about the first and remarkably patient about the second, whereas anxious leaders tend to blur the two together and end up leaning hardest on the one kind of lag that should be left alone.

In practical terms, that means an update along the lines of “we don’t know yet, here’s what we’re doing about the immediate risk, and you’ll hear from us again at four o’clock” ought to be entirely acceptable. When a team commits to a rhythm of communication rather than promising certainty it doesn’t yet have, the leader gets the reassurance they legitimately need and the team keeps the space it needs to think. Not every gap in communication is a failure of communication, and quite often it’s just the room required to get the answer right.

7. The Control Paradox: Why Anxious Leaders End Up Less Informed

The cost in hours is real, but I don’t think it’s the most dangerous part of the tax. The more serious problem is what anxious leadership does to the quality of the information a leader receives, because the very behaviour that’s meant to keep them informed can end up steadily degrading what they hear.

The mechanism isn’t complicated once you see it. When every update attracts scrutiny and every ambiguous status triggers a meeting, people quite rationally start managing the leader’s reactions rather than the underlying problem. They report what’s safe to report, they escalate what’s politically acceptable to escalate, and they learn to round uncertainty up into confidence because confidence is what gets rewarded. The messy, half finished and slightly awkward observations that are so often the first sign of real trouble stop travelling upwards, and what arrives instead is a cleaner, greener and far less accurate picture of reality.

From the leader’s chair, this looks like success. The dashboards are green, the updates are tidy and the surprises seem to have stopped, so it feels as though the extra governance is working. That false confidence is exactly what makes the loop so hard to break, because when reality eventually does break through, usually as an incident or a missed delivery that nobody saw coming, the natural response is to conclude that there still isn’t enough oversight and to ask for more of it. The anxiety that started the cycle ends up being fed by the cycle itself.

%%{init: {'theme':'base','themeVariables':{'primaryColor':'#1f2937','primaryTextColor':'#ffffff','primaryBorderColor':'#f59e0b','lineColor':'#f59e0b','fontSize':'16px'}}}%%
flowchart TD
    A[Leadership anxiety] --> B[More reporting and scrutiny]
    B --> C[Defensive behaviour]
    C --> D[Sanitised information]
    D --> E[False confidence]
    E --> F[Reality breaks through as a surprise]
    F --> A
    classDef node fill:#1f2937,stroke:#f59e0b,color:#ffffff,stroke-width:2px;
    class A,B,C,D,E,F node;

This is the watermelon project at an organisational scale, green on the outside and red all the way through, except that nobody is deliberately hiding anything; everyone is simply responding sensibly to the incentives in front of them. The uncomfortable conclusion is that the more a leader demands certainty, the more likely they are to receive confidence instead of truth, and the better informed they feel, the less informed they may actually be.

8. What the Research Does, and Does Not, Tell Us

None of this is just my own frustration talking, although I want to be careful about what the research actually supports. There’s a solid body of organisational work on the individual mechanisms involved, but none of it shows that anxious leadership inevitably curdles into something worse, and I’d rather make the narrower claim that holds up than the dramatic one that doesn’t.

Julia DiGangi, a neuropsychologist, wrote a piece for Harvard Business Review in 2023 called The Anxious Micromanager, drawing on the ideas in her book Energy Rising. It’s a practitioner essay rather than an empirical study, but it makes the useful argument that a great deal of controlling behaviour, such as excessive checking, questioning and stepping in, can be understood as leaders managing their own anxiety rather than responding to anything the team actually needs. I find that framing helpful because it explains how a leader can sincerely say they want an independent, empowered team while behaving in ways that make independence almost impossible.

Amy Edmondson’s 1999 study of 51 work teams in a manufacturing company found that psychological safety, which she defined as a shared belief that the team is safe for interpersonal risk taking, was associated with learning behaviour such as seeking feedback, talking openly about errors and asking for help, and that this learning behaviour was in turn associated with how well teams performed. It’s a correlational field study, so it doesn’t prove that controlling leaders cause teams to stop learning, but it does suggest that places where people feel unsafe admitting uncertainty are likely to be places where less learning goes on.

Elizabeth Morrison and Frances Milliken’s 2000 paper on organisational silence offers a conceptual model rather than a dataset. They proposed that certain conditions, including senior management’s fear of negative feedback and a set of implicit beliefs that employees are self interested and that agreement is healthy while dissent is not, can lead people to conclude collectively that raising problems is either pointless or risky. Morrison’s later review of the employee voice literature, published in 2014, reinforced the broader point that people make quite deliberate calculations about whether speaking up is worth the personal cost, and that organisations lose a great deal of valuable information whenever the answer comes out as no. Of all the work cited here, these two pieces map most directly onto the control paradox in the previous section.

Colin Fisher, Teresa Amabile and Julianna Pillemer also wrote in Harvard Business Review in 2021 about how leaders can help their teams without micromanaging them, drawing on their own research into how help is given and received inside organisations. Their central argument is that well meant intervention, offered at the wrong moment or without being clear about the role you’re playing, can undermine people’s sense of autonomy and damage the relationship even when the intention behind it is entirely good.

Taken together, these don’t add up to a law, but they’re consistent with the pattern I’ve described: when speaking up feels risky and intervention feels constant, the information that reaches the top gets thinner, and the leader is usually the last to notice.

9. When Anxiety Spreads Across the Executive Team

So far I’ve written as though this is a story about one leader and their team, but in practice the tax rarely stays in one reporting line. Anxiety travels sideways across an executive committee just as easily as it travels down. One executive who can’t tolerate uncertainty about another function’s work starts asking that function for its own trackers and assurance packs, the other function responds by building defensive governance of its own, and before long two or three teams are producing reports for each other mainly to prove that they can’t be blamed. What started as one person’s discomfort becomes friction between whole divisions, and the organisation ends up spending its energy on internal reassurance rather than on its customers.

Part of what drives this, if we’re honest, is that some leaders confuse being informed with being important. Every reporting line, mandatory update and meeting that can’t start without them is a small reminder of where they sit in the hierarchy, but the amount of information flowing towards a leader is not a measure of their effectiveness; quite often it’s just a measure of how much of the organisation’s attention they consume. The most expensive person in an organisation may not be the one making the biggest decisions, but the one who needs the most reassurance before anyone else can make theirs.

10. When the Fear of Surprise Becomes Dangerous

Which brings me back to the documentary. I don’t think leaders who dislike surprises are tyrants, and it would be both absurd and unfair to suggest it, because most of them are simply people carrying the weight of a lot of responsibility as best they can. What I do think is that there’s a continuum here that’s worth understanding, precisely because the slide along it is so gradual that almost nobody notices it while it’s happening.

It tends to start with an entirely reasonable wish to know what’s going on, which over time hardens into an expectation of constant updates. Those updates eventually become mandatory reporting, the reporting invites scrutiny, the scrutiny leads to intervention, and enough intervention can leave you with a culture in which people are reluctant to do anything without first asking permission. Nobody set out to build that culture and nobody ever stood up and announced that independent thinking was no longer welcome, but the organisation learned it all the same, one interaction at a time, until avoiding surprising the person in charge had become the safest thing anyone could do.

That’s where the comparison with tyranny earns its place, not as an accusation but as a warning about direction of travel. The real danger isn’t simply that a leader wants control; it’s that everybody else gradually loses the freedom to act on their own judgement, and when that happens something quite precious tends to disappear along with it, which is the chance of ever being pleasantly surprised.

11. Trust Is Not the Absence of Accountability

One objection I hear fairly often is that reducing reporting means reducing accountability, and I think that rests on a misunderstanding of what accountability really is. Being accountable means being clear about outcomes, responsibilities, standards and consequences; it doesn’t mean producing a continuous stream of evidence that you’re busy.

A team can be highly accountable while having a great deal of autonomy, and in my experience autonomy usually strengthens accountability rather than weakening it, because people who genuinely own a decision feel responsible for how it turns out in a way that people who are merely carrying out instructions rarely do. So the more useful question isn’t whether a leader knows everything their teams are doing, but whether those teams understand what they’re trying to achieve, have the freedom to go and achieve it, and know when to ask for help. When those things are in place, leadership becomes much less about supervising activity and much more about creating the conditions for excellent work, and that asks something of leaders that turns out to be surprisingly hard, which is restraint.

12. Practical Ways to Lower the Tax

All of this is only worth writing about if it changes what we actually do on a Monday morning, so I’ll finish the argument with the things I’ve seen make the biggest difference.

The first is to audit your recurring reports properly. Write down every regular update, dashboard and meeting that you receive or chair, ask of each one when it last changed a decision, and retire the ones that can’t give you a good answer. It’s worth telling the team when you’ve done it, because in my experience they won’t quite believe the reports have gone until they see it in writing.

The second is to make your escalation thresholds explicit, which means spelling out what must reach you immediately, what can wait for the normal rhythm and what you never need to hear about at all. Vague thresholds are one of the biggest drivers of over reporting, because when nobody is sure what matters to you, the safest thing anyone can do is send you everything.

The third is to keep escalation lag and understanding lag firmly apart, insisting on speed when it comes to raising material risks while protecting the time teams need to work out what’s really going on. During incidents, that usually means agreeing a cadence of communication up front rather than demanding answers that don’t exist yet.

The fourth is to make a visible point of thanking whoever brings you bad news early, especially when it’s incomplete or turns out to be less serious than it first looked. People watch very closely what happens to the first person who speaks up, and how you react in that moment will shape what you hear for years afterwards.

The last one is the most personal, which is simply to notice your own urge to check in. Before asking for an update, it’s worth pausing to ask yourself whether you need the information to make a decision or just to feel a bit better, and if it’s the second of those, the most useful thing you can do for your team is usually to wait. If you only take one question away from this post, make it this one: how much of my organisation’s time is spent solving real problems, and how much is spent making me feel comfortable?

13. Leave Room to Be Surprised

I still think about the leader who told me never to surprise him. I wonder how many good ideas never made it as far as his office, how many people simply stopped challenging him, how many opportunities slipped away because someone decided it was safer to stay on the expected path, and how much of the organisation’s energy went into keeping one person comfortable. I also wonder how differently things might have turned out if he’d said something else that first week, perhaps something like “never hide a risk from me, but do surprise me with what you find, push back on my assumptions, and come to me when you need help rather than when I need reassurance.”

One of the real privileges of leadership is getting to work with talented people who can do things you’d never have thought of yourself, people who know things you don’t, who spot the opportunities you walked straight past and who solve problems in ways that wouldn’t have occurred to you. If every outcome has to be anticipated, every decision approved and every development reported before it happens, it’s fair to ask what room is actually left for those people to contribute anything of their own.

Perhaps the real measure of a leader’s maturity isn’t how much uncertainty they manage to squeeze out of their organisation, but how much of it they’re willing to carry themselves without passing the anxiety down to everyone else. Because when leaders make it impossible to surprise them, they really shouldn’t be surprised when their organisations stop doing anything remarkable.

References and Further Reading

Julia DiGangi (2023), The Anxious Micromanager, Harvard Business Review, 101(5), 52 to 55, September to October 2023. Practitioner essay; see also her book Energy Rising: The Neuroscience of Leading with Emotional Power (Harvard Business Review Press, 2023).
Amy Edmondson (1999), Psychological Safety and Learning Behavior in Work Teams, Administrative Science Quarterly, 44(2), 350 to 383. Field study of 51 work teams.
Elizabeth Wolfe Morrison and Frances J. Milliken (2000), Organizational Silence: A Barrier to Change and Development in a Pluralistic World, Academy of Management Review, 25(4), 706 to 725. Conceptual model.
Colin M. Fisher, Teresa M. Amabile and Julianna Pillemer (2021), How to Help (Without Micromanaging), Harvard Business Review, 99(1), 123 to 127, January to February 2021.
Elizabeth Wolfe Morrison (2014), Employee Voice and Silence, Annual Review of Organizational Psychology and Organizational Behavior, 1, 173 to 197. Literature review.