.png)
Trust Wins: What the Research Says About Micromanagement, Autonomy and the Space to Do Good Work.
Updated: Sep 28
The message arrives at eight minutes to five. Where are we on the board pack? It is the third time today, you sent a full update at lunch, and the pack is in the shared folder with a note on the one slide still waiting for finance. You answer politely, attach the link again, and feel something quietly drain out of the afternoon. The work is fine. What has gone is the sense that anyone thinks you can be left to do it.
I have worked for both kinds of leader: a manager whose favourite questions were "Have you done this?" and "Where is that?", and a leader who trusted me to get on with it and, when something worried them, asked how they could help. The difference felt enormous at the time, and it turns out the research agrees, though it has a few things to say that are more interesting than "trust good, control bad". Close watching has costs that land on both people, including the one doing the watching. Autonomy meets a real human need. Structure, handled well, is a kindness rather than an insult. And handing people the reins can go wrong in ways that look a lot like being abandoned. This post is that research, and what it means for you whether you are the one being checked on or the one doing the checking.
What you will get out of this
What 94 studies and more than 23,000 workers say electronic and close monitoring does to performance and stress, and the famous experiment on the hidden costs of control, with the replications that softened it
The strange thing close watching does to the watcher, first found in 1958 and found again in a trust game sixty years later
Why the same check-in can feel like support or suspicion, and the three features that turn attention into micromanagement
What autonomy does, from meta-analyses covering more than 440,000 people, and the training study that made managers more autonomy-supportive in five weeks
The old question of structure versus consideration, and why the answer is "both, in the right amounts"
When empowering leadership backfires: the teams that did better under direction, the burden of being trusted, and the thin line between empowering and absent
How to work well for someone who checks everything, and how to avoid becoming that person when you manage others
The check-in check: a tool that reads your situation and drafts a reporting rhythm you can propose
The Trust Pages: three pages for building the record, agreeing the rhythm and handing work over well
What close watching does, on both sides of the desk
Start with the largest body of evidence, because it is the least ambiguous. Electronic performance monitoring, meaning any technology that observes, records or analyses information about someone's work, has been studied for more than forty years, and in 2023 a team of researchers pooled 94 independent samples covering 23,461 workers. They found no evidence that monitoring improved performance. They did find that its presence went with more stress, whatever the characteristics of the monitoring, and that organisations which monitored more transparently and less invasively got more positive attitudes from the people being monitored. A separate meta-analysis came to the same shape of answer: small costs to job satisfaction and stress, and no relationship with performance. A 2025 review of the whole field suggested why the performance effect sits so stubbornly at zero, which is that monitoring seems to push people in two directions at once and the two pushes cancel out.
The laboratory work shows who gets pushed which way. In a 1995 experiment on a data-entry task, highly skilled people who were monitored keyed more entries than highly skilled people who were not, and the opposite happened for people with less skill. Being watched made the confident faster and the unsure slower. The people who felt least stressed were the ones nobody was monitoring and the ones working in a close-knit group.
Then there is the experiment most people quote. In 2006 two economists ran a game in which one person could choose to impose a minimum performance requirement on another before the second person decided how hard to work. Most of the people who were controlled reduced their effort in response, and when asked why, most said they read the control as a signal of distrust and a limit on their choice. The researchers called it the hidden costs of control. It deserves a footnote, and the research has supplied one. Four repetitions with 476 participants confirmed that the hidden costs exist, and found them usually too small to undo the benefit of the control itself. Theory work around the same time explained the mechanism: choosing to control someone reveals what you expect of them, and people respond to the expectation. A further experiment found that people who had first built something together as a team reacted to control even more sharply, which will surprise nobody who has had a trusted working relationship interrupted by a sudden request for daily status reports.
Watching someone closely does not only change how they work. It changes what you are able to learn about them.
Now the finding that I think matters most, and it concerns the person doing the watching. In a study published in 1958, supervisors oversaw two workers, one monitored closely and one left alone, whose performance was identical. The supervisors came to trust the monitored worker less, because they attributed that worker's good performance to the monitoring rather than to the person. Sixty years later a set of trust games found the same trap from another angle. People behaved well when they knew they were being monitored and took advantage when they knew they were not, and the people doing the monitoring mistook the compliance they saw for trustworthiness, so they misplaced their trust the moment they stopped watching. Close watching destroys the very evidence the watcher needs in order to stop.
There is a field version of this too. At one of the largest mobile phone factories in the world, a researcher found that workers who were fully visible to supervisors hid their small improvements and workarounds, because anything unusual drew correction. Giving production lines a modest zone of privacy, a curtain around the line, sustainably and significantly improved their performance. The researcher called it the transparency paradox: making people more observable can make them less productive, because they spend effort looking right instead of working well.
And the honest counterweight, because monitoring is sometimes exactly what a situation needs. A randomised policy experiment in Sweden replaced a trust-based contract between a foreign aid agency and the non-profit organisations it funded with closer monitoring. Rather than crowding out motivation, it increased the organisations' outreach, reduced their expenditure and cut the number of financial irregularities. Where the stakes are public money and the relationship is at arm's length, oversight did its job. The research on monitoring does not say that checking is wrong. It says that checking by default, without transparency or purpose, costs more than it returns, and that it quietly stops the checker from ever finding out whether the checks are needed.
Try this. Next time you are about to check on someone, or you are the one being checked, ask what the check is for. If it is to catch a problem early, say so and agree when the checks will stop. If nobody can say what it is for, it is probably a habit, and the 1958 finding says habits of watching tend to feed themselves.
The research behind this section
Ravid and colleagues (2023) meta-analysed 94 independent samples (N = 23,461) and found no evidence that electronic performance monitoring improved worker performance, found that its presence was associated with increased worker stress regardless of the characteristics of the monitoring, and found that organisations which monitored more transparently and less invasively could expect more positive worker attitudes. Ravid and colleagues (2020) reviewed the monitoring literature since 2000 and proposed a typology of monitoring characteristics to explain contradictory findings. Siegel, König and Lazar (2022) meta-analysed the effects of electronic monitoring and found small negative effects on job satisfaction, small positive effects on stress and no relationship with performance. König (2025) reviewed research on electronic monitoring at work and found the overall impact on performance neutral, with a small positive correlation with strain and a small negative correlation with job attitudes, possibly because paradoxical effects counterbalance each other. Aiello and Kolb (1995) found in a laboratory data-entry task that highly skilled monitored participants keyed more entries than highly skilled unmonitored participants while the opposite pattern appeared among low-skilled participants, and that unmonitored workers and members of cohesive groups felt the least stressed. Kolb and Aiello (1996) found that locus of control moderated the effect of monitoring on stress, with internals more stressed when monitored and externals more stressed when not. Falk and Kosfeld (2006) found in a principal-agent experiment that most agents reduced their performance when the principal chose to impose a minimum requirement, and that most who reacted negatively perceived control as a signal of distrust and a limitation of their autonomy. Ziegelmeyer, Schmelz and Ploner (2012) ran four repetitions with 476 participants and largely confirmed the hidden costs of control, but found them usually not substantial enough to undermine the effectiveness of the incentives. Sliwka (2007) set out a model in which the incentives or autonomy a principal chooses reveal the principal's beliefs about the prevailing social norm, so that control can crowd out motivation by conveying pessimism. Riener and Wiederhold (2016) found that agents who had first gained common experience in a team-building task outperformed others but had their effort crowded out more strongly by control. Frey and Jegen (2001) surveyed the empirical evidence on motivation crowding. Strickland (1958) found that supervisors who monitored one of two equally performing workers more closely came to trust that worker less, attributing their compliance to the surveillance. Schweitzer, Ho and Zhang (2018) found in repeated trust games that trustees behaved compliantly when they anticipated monitoring and exploited trustors when they did not, and that trustors misattributed the compliance they observed to trustworthiness and so misplaced their trust when they could not monitor. Bernstein (2012) used embedded observers and a field experiment at a large mobile phone factory in China and found that even a modest increase in group-level privacy sustainably and significantly improved line performance, supporting productive deviance, localised experimentation and continuous improvement. Bengtsson and Engström (2014) ran a randomised policy experiment with Swedish non-profit organisations, replacing a trust-based contract with increased monitoring, and found that the intervention increased outreach, reduced expenditures and reduced financial irregularities.
Why the same question can feel like support or suspicion
"How is it going?" can be the kindest question in the working week or the most deflating, and the research has started to explain the difference.
The first distinction is between watching and talking. A 2016 study of 388 supervisors and their direct reports separated two styles of monitoring: observational, where the supervisor keeps an eye on the work from a distance, and interactional, where the supervisor checks in through conversation. The two styles had opposite relationships with trust and distrust, which flowed through into the quality of the working relationship, whether people asked for feedback, and in the end how innovative their supervisors rated them. Being looked at and being talked to are experienced as different acts, even when the information gathered is the same.
The second distinction is predictability. Two diary studies of hybrid workers, with more than 3,600 daily reports between them, found that on days when supervisors monitored more, people felt less trusted, and on those days they were more exhausted and had less energy. Where monitoring was erratic, high one day and absent the next, the dent in felt trust was deeper, at least in teams that had newly moved to remote work. An unpredictable check reads as suspicion far more than a regular one.
The third distinction is purpose. Among 3,744 public health doctors working remotely, monitoring that felt controlling went with higher burnout, and monitoring that felt developmental went with lower burnout. An experiment and a field study published in 2023 found that electronic monitoring weakened the social exchange between supervisors and employees, which led to more cutting of corners and lower task performance, and that supervisors who gave the monitoring data back to employees as developmental feedback buffered most of that damage. A companion study found that monitoring also blunted the trust-building effect of otherwise ethical leaders. The data itself was neutral. What people reacted to was whether it came back to them as help or disappeared upwards as evidence.
Which brings us to the word itself. For a term everyone uses, micromanagement had never been properly defined or measured until a series of studies published in 2025, across eight samples and 1,723 people. The researchers found three core attributes, controlling, close monitoring and a focus on detail, and three features that turn those attributes into micromanagement: they are excessive, sustained and unnecessary. Micromanagement went with low trust and a weaker working relationship, and with higher intentions to leave and more emotional exhaustion, and it was distinct from neighbouring ideas such as directive leadership and initiating structure.
Attention becomes micromanagement when it is more than the work needs, goes on longer than it needs to, and is aimed at things that did not need checking.
That definition leaves room for close attention that is proportionate, time-limited and aimed at what matters. A new starter's first fortnight or a high-stakes launch can warrant detail without any of it being micromanagement. Under the micromanager I worked for, the endless check-ins came across as doubt rather than accountability, and the unspoken team motto became "get it done, so they will leave us alone". That is what the research would predict when checking is excessive, sustained and unnecessary all at once.
Try this. If you are being checked on, notice which of the three features is present. If the checking is excessive, offer a single fuller update in place of several small ones. If it is sustained, suggest a date to review it. If it is aimed at the wrong things, say which parts carry the risk, so that attention goes there.
The research behind this section
Liao and Chun (2016) developed scales for observational and interactional supervisor monitoring and, in a survey of 388 supervisor-subordinate dyads, found that the two styles were related to subordinates' trust and distrust in their supervisor, which were in turn related to leader-member exchange quality, feedback-seeking behaviour and supervisor-rated innovation, so that the effect of monitoring on innovation depended on the type of monitoring. Zheng and colleagues (2023) conducted two experience sampling studies of hybrid workers (N = 191 with 1,417 data points and N = 257 with 2,244 data points) and found that daily supervisor monitoring was negatively associated with daily felt trust, which in turn reduced daily well-being, and that monitoring variability intensified the negative relationship between monitoring and felt trust in a newly introduced remote working context but not in a more stable one. Miao and colleagues (2025) studied 3,744 remote-working public health doctors and found that monitoring perceived as controlling led to higher burnout through the psychological contract, while monitoring perceived as developmental led to less burnout. Thiel and colleagues (2023) found across an online experiment and a field study that electronic monitoring undermined leader-member social exchange, increasing production deviance and reducing task performance, and that supervisors who gave monitoring data back to employees as developmental feedback buffered these consequences. Thiel, Prince and Sahatjian (2022) found in multi-wave field data that electronic monitoring undermined the ability of ethical leaders to build trust with their employees. Deen and colleagues (2025) defined and validated a micromanagement scale across eight samples of 1,723 people and found that micromanagement comprised three core attributes (controlling, close monitoring and detail focus) and three key features (excessive, sustained and unnecessary), that it related to low trust and leader-member exchange and to higher turnover intentions and emotional exhaustion, and that it was distinct from authoritarian, participative, empowering and directive leadership, initiating structure and abusive supervision.
The need underneath: autonomy, and what it does
Why does being checked on so often feel worse than the time it takes? Self-determination theory, one of the most tested theories in the study of motivation, has a clear answer: people have three basic psychological needs at work, for autonomy, competence and relatedness, and work that meets them is done with more energy, more care and less strain. Autonomy here does not mean independence or working alone. It means feeling that you endorse what you are doing, that it is in some real sense yours.
The evidence for the link is now very large. A 2016 meta-analysis of 99 studies found that satisfying these needs at work was associated with growth, internalisation of the work's value and well-being. A 2018 meta-analysis focused on leaders, pooling 72 studies and 32,870 people, found that when leaders supported autonomy, their people were much more likely to be motivated for their own reasons, and no more likely to be motivated by pressure or reward; autonomy support also went with better well-being, less distress and more positive behaviour at work, and none of this varied by country. A 2024 meta-analysis went further still, with 881 samples and 443,556 people. Supporting all three needs was strongly related to their satisfaction and to well-being, and had small to moderate positive associations with performance.
That last phrase is worth sitting with, because it is the honest version of the claim. Autonomy support does not transform performance on its own; its strongest effects are on how people feel and why they work. The older evidence says the same. A 1986 meta-analysis found that employees who felt more control over their work were more satisfied, committed, involved and motivated, performed somewhat better, and reported fewer physical symptoms, less distress and less absence. A study in two workplaces found that managers' support for autonomy predicted whether employees' needs were met, which in turn predicted both their performance ratings and their psychological adjustment.
The 2024 meta-analysis contains one finding that complicates the usual story, and it is worth knowing if you work across cultures. The link between autonomy support and people's own motivation was weaker in more individualist cultures. That is the opposite of what many people would guess, and it suggests that autonomy is valued everywhere, though not always expressed or received in the same way.
Then the most practical question: can a controlling manager change? The evidence says yes, more readily than you might think. In 1989 a study in a large corporation found that managers' support for autonomy went with employees' satisfaction and trust, and that an intervention could shift the managers' orientation. In a 2009 experiment, 25 managers at a Fortune 500 company were trained in how to support autonomy. Five weeks later they were measurably more autonomy-supportive than untrained managers, and the 169 employees they supervised showed more self-motivation and greater engagement. A 2021 systematic review of ten such interventions, three of them randomised trials, found that seven produced mostly favourable effects, two mixed and one none. It also found that change showed up more reliably in the managers than in their teams, which suggests patience is part of the method.
Why are some managers controlling in the first place? One answer, from a study of teachers, is uncomfortable and illuminating. The teachers who felt most pressure from above, from curricula, colleagues and performance standards, became less self-motivated themselves and more controlling with their students. Pressure travels downwards. A manager who checks everything is very often a manager who is being checked on, and that is one of the most useful things to know if you work for one.
Try this. Look at one piece of your work that you do with the most energy, and one you do with the least. Ask whether the difference is the task or how much say you have over it. If it is the say, the fix may be as small as agreeing the outcome and choosing the method yourself.
The research behind this section
Deci, Olafsen and Ryan (2017) reviewed self-determination theory in work organisations. Gagné and Deci (2005) set out self-determination theory as a theory of work motivation, distinguishing autonomous from controlled motivation. Van den Broeck and colleagues (2016) meta-analysed 99 studies with 119 samples on the antecedents and consequences of basic need satisfaction at work. Slemp and colleagues (2018) meta-analysed 754 correlations from 72 studies (83 samples, N = 32,870) and found that leader autonomy support correlated strongly and positively with autonomous work motivation and was unrelated to controlled motivation, and was positively associated with basic needs, well-being and positive work behaviours and negatively associated with distress, with no moderation by country, source of support or publication status. Slemp and colleagues (2024) meta-analysed 4,561 effect sizes from 881 independent samples (N = 443,556) and found that supports for autonomy, competence and relatedness were strongly related to need satisfaction and subjective well-being and had small to moderate positive associations with performance, and that the correlation of autonomy support with autonomous motivation weakened as individualism increased. Spector (1986) meta-analysed studies of perceived control, covering autonomy and participative decision-making, and found high perceived control associated with higher satisfaction, commitment, involvement, performance and motivation, and with lower physical symptoms, emotional distress, role stress, absenteeism, intent to turnover and turnover. Humphrey, Nahrgang and Morgeson (2007) meta-analysed motivational, social and contextual work design features, including autonomy. Baard, Deci and Ryan (2004) found in two work organisations that perceived manager autonomy support and employees' autonomy orientation predicted need satisfaction, which predicted performance evaluations and psychological adjustment. Deci, Connell and Ryan (1989) found in a large corporation that managers' support for self-determination related to employees' satisfaction and trust, and tested an organisational intervention. Hardré and Reeve (2009) trained 25 managers at a Fortune 500 company in autonomy support and found five weeks later that they displayed a more autonomy-supportive style than untrained managers, and that their 169 employees showed more autonomous motivation and greater workplace engagement. Slemp, Lee and Mossman (2021) reviewed ten field interventions (three randomised controlled trials, combined N = 2,337) designed to develop need-supportive climates or leader behaviours and found seven with mostly favourable effects, two mixed and one with no change, with change more evident at the leader level than the subordinate level. Pelletier, Séguin-Lévesque and Legault (2002) found that teachers who perceived more pressure from above and from below were less self-determined and more controlling towards their students.
Structure and consideration, and what each is for
The micromanager I worked for was a textbook case of what the leadership literature calls initiating structure: roles, tasks, timelines, process. The leader who trusted me was high on what it calls consideration: warmth, respect and attention to the person. These two ideas come from the Ohio State studies of the 1950s and were, for a while, treated as rather old-fashioned. Then in 2004 a meta-analysis looked at them again.
It pooled 163 correlations for consideration and 159 for initiating structure, and found that both had moderately strong relationships with leadership outcomes. Consideration was the stronger of the two overall, at .48 against .29 for structure. But the detail is what matters. Consideration was more strongly related to how satisfied and motivated people were and how effective they rated the leader. Initiating structure was slightly more strongly related to the leader's own performance and the performance of the group or organisation. The two are doing different jobs. A later meta-analysis of traits and behaviours together found that what leaders do explains more of their effectiveness than who they are, and that both of these behaviours earn their place.
So structure is not the villain of the story. What the research on fit adds is that the amount matters, and it matters differently for each. A 2012 study compared the structure and consideration people said they needed with what they received. Too little of either was bad for trust, satisfaction and commitment. Too much consideration was still received well. Too much structure was not. People forgive a leader for caring more than they strictly need; they do not forgive being organised more than they strictly need.
The team research tells a similar story with a twist. A meta-analysis of leadership in teams found task-focused behaviours related to perceived team effectiveness and productivity, and person-focused behaviours related to both a little more strongly, and very strongly indeed to team learning. Empowerment behaviours alone accounted for nearly 30% of the variance in how much teams learned. Direction gets the work done; involvement gets the team better at doing it.
There is an older idea that complicates this helpfully. In 1978 two researchers proposed that some features of the work, the people and the organisation act as substitutes for leadership: an experienced professional with clear feedback from the task itself needs far less structure from a manager, and a routine, well-documented process needs less direction. A 1996 meta-analysis found that these substitutes explained a good deal of employees' attitudes and performance in their own right. It is a useful reminder for anyone in your role. Much of what you build, the trackers, the templates, the standing arrangements, is structure that does not need the manager to supply it, and when it exists, a manager has less reason to supply it themselves.
What about the popular idea that leaders should simply read each person's readiness and adjust their style to match? Situational leadership is one of the most widely taught models in management training, and a 2009 test of three versions of it found little support for its detailed prescriptions. The broad intuition, that people new to a task need more structure than people who have mastered it, is sound and appears elsewhere in this research. The precise matrix taught on courses is less well supported than its popularity suggests.
Try this. Write down the two or three pieces of structure you most need from your manager, such as a decision deadline, an agreed priority or a single point of contact, and the two or three you would happily do without. Offer the first list and quietly build the structure yourself for the rest. Asking for the structure you need is a fine way of making the structure you do not need unnecessary.
The research behind this section
Judge, Piccolo and Ilies (2004) meta-analysed 163 correlations for consideration and 159 for initiating structure and found that both had moderately strong, nonzero relations with leadership outcomes (consideration .48, initiating structure .29), with consideration more strongly related to follower satisfaction, motivation and leader effectiveness and initiating structure slightly more strongly related to leader job performance and group-organisation performance. DeRue and colleagues (2011) integrated trait and behavioural theories of leadership and found that leader traits and behaviours together explained a minimum of 31% of the variance in leadership effectiveness criteria, with behaviours tending to explain more than traits. Judge and Piccolo (2004) meta-analysed transformational and transactional leadership and their relative validity. Lambert and colleagues (2012) found that deficient amounts of both consideration and initiating structure were associated with unfavourable attitudes, that excess consideration was associated with favourable attitudes while excess initiating structure was associated with unfavourable ones, and that fit between needed and received leadership influenced citizenship behaviour partly through trust in the supervisor. Burke and colleagues (2006) meta-analysed leadership behaviours in teams and found task-focused behaviours related to perceived team effectiveness (.333) and productivity (.203) and person-focused behaviours related to perceived effectiveness (.360), productivity (.284) and team learning (.560), with empowerment behaviours accounting for nearly 30% of the variance in team learning. Kerr and Jermier (1978) proposed that characteristics of subordinates, tasks and organisations can substitute for or neutralise leadership. Podsakoff, MacKenzie and Bommer (1996) meta-analysed the substitutes for leadership and found that they accounted for substantial variance in employee attitudes and performance in their own right. Thompson and Vecchio (2009) tested three versions of situational leadership theory and found little support for its prescriptions.
When handing over the reins goes wrong
Empowering leadership, meaning a leader who shares power, explains why the work matters, involves people in decisions and expresses confidence in them, has a strong evidence base. A 2018 meta-analysis of 105 samples found positive effects on performance, going beyond the job and creativity, at both individual and team level, working through trust in the leader and people's sense of being empowered. A second 2018 meta-analysis of 55 samples found the strongest link was with how people felt about the leader, a correlation of .59, and a weaker but real link with behaviour and performance, .31. It found no relationship at all with people's emotions. Empowering leaders are liked more reliably than they are followed to better results.
The research of the last decade has become interested in when it goes wrong, and it has found at least five ways.
It can be too much. A study of 137 employees and their supervisors found an inverted U: performance rose with empowering leadership up to a point and then fell, and the peak depended on whether the person enjoyed learning. A 2024 team study found the same shape. A 2016 study named the mechanism: empowering behaviour enabled people through greater confidence, and also burdened them through job-induced tension, and the burden ate into the benefit.
It can arrive too early. In a study of 60 teams, teams with a directive leader outperformed teams with an empowering leader at first. Over time the empowered teams improved faster, because they learned more, coordinated better and built a better shared picture of the task, and overtook them. That is a wonderful finding for the patient. But a 2020 experiment with 67 teams in a firefighting simulation found directed teams outperformed empowered ones both before and after a sudden change in the task, which is exactly where the theory expected empowered teams to shine. The two studies disagree, and my reading is that time is the variable: empowerment pays off across a long run of learning, and direction wins in short, high-pressure stretches.
It can help the wrong people least. A study of sales representatives found that empowering leadership did most for the people with the least product knowledge and experience, and little for the most experienced, who presumably had the autonomy they needed already. A field experiment in work units in the United Arab Emirates found that both directive and empowering leadership improved core task proficiency, and only empowering leadership increased proactive behaviour. A study of 140 school teams found directive leadership went with commitment and day-to-day performance, and participative leadership with empowerment and innovation. Each style bought something different.
It can be read as absence. Here is the finding every well-meaning manager should know. In a study of 150 leaders and their followers, empowering behaviour that did not match what the follower expected, whether more or less, was perceived as laissez-faire leadership, and those leaders were rated as less effective. A related study found that delegation itself could be perceived as laissez-faire, depending on how competent the leader seemed. And laissez-faire leadership is not benign. In a survey of 2,273 employees in Norway, it went with role conflict, role ambiguity and conflict with colleagues, and through them with bullying and distress. A three-wave study over five years found it predicted role ambiguity at each point, while initiating structure and consideration did not. Leaders who simply failed to respond to good or poor performance were rated less effective and their people less clear about their roles.
It needs feedback to point it somewhere. In a study of 110 teams in a manufacturing firm, the highest-performing teams combined high autonomy with frequent performance feedback, because they understood the organisation's goals. The lowest-performing teams had high autonomy and little feedback. A study of self-managing teams found that very high trust could make team members reluctant to monitor one another at all, and that high trust combined with high individual autonomy could hurt performance. Autonomy without information about how it is going is closer to guesswork than to trust.
Empowerment without feedback is not freedom. From the inside, it feels a lot like being left to it.
There is one more, quieter risk. A study of 74 teams found that when a leader empowered some team members much more than others, the team shared less information and performed worse. Trust handed out unevenly becomes a status marker.
None of this undoes the case for empowering people. It sharpens it. The research reviews in this area now argue that empowering leadership works through two routes at once, confidence and burden, and that its effects depend on the person, the timing and whether anyone is telling people how it is going. The leader I loved working for got this right without, I suspect, having read any of it: the trust came with a standing offer of support and a clear sense of what good looked like.
Try this. When work is handed to you with "you decide", ask two questions before you start: what does good look like, and when should I show you where I have got to? The first question turns freedom into direction. The second turns it into channelled autonomy, which is the version that performed best.
The research behind this section
Lee, Willis and Tian (2018) meta-analysed 105 samples and found positive effects of empowering leadership on performance, organisational citizenship behaviour and creativity at individual and team levels, with trust in the leader and psychological empowerment mediating at the individual level, and incremental contribution over transformational leadership and leader-member exchange. Kim, Beehr and Prewett (2018) meta-analysed 55 samples and found the strongest correlation between empowering leadership and attitudes towards the leader (ρ = .59) and the weakest with behavioural and performance outcomes (ρ = .31), with no significant relationship with emotions. Seibert, Wang and Courtright (2011) meta-analysed the antecedents and consequences of psychological and team empowerment. Lee and colleagues (2017) found in 137 supervisor-subordinate dyads an inverted U-shaped relationship between empowering leadership and task performance, moderated by learning orientation. Chen and colleagues (2024) found a curvilinear, inverted U-shaped effect of empowering leadership on team performance through team reflexivity. Cheong and colleagues (2016) found that empowering leadership enabled followers through self-efficacy and burdened them through job-induced tension, which diminished its positive influence on work role performance. Cheong and colleagues (2019) reviewed the mixed results on empowering leadership and proposed examining non-linear effects, reverse causation, contradictory mediating mechanisms and boundary conditions. Sharma and Kirkman (2015) reviewed the empowering leadership literature and set an agenda for studying its unintended negative outcomes. Lorinkova, Pearsall and Sims (2013) found in longitudinal data from 60 teams that teams led by a directive leader initially outperformed those led by an empowering leader, but that empowered teams improved more over time through team learning, coordination, empowerment and mental model development. Sanchez-Manzanares and colleagues (2020) found in an experiment with 67 teams in a firefighting simulation that directively led teams outperformed empowered teams both before and after a task change. Ahearne, Mathieu and Rapp (2005) found that empowering leader behaviours were most beneficial for sales representatives with low product and industry knowledge and low experience. Martin, Liao and Campbell (2013) found in a field experiment in the United Arab Emirates that directive and empowering leadership both increased core task proficiency but only empowering leadership increased proactive behaviour, with boundary conditions depending on units' satisfaction with their leaders. Somech (2005) found in 140 school teams that directive leadership related to commitment and in-role performance and participative leadership to empowerment and innovation. Wong and Giessner (2018) found in a two-stage study of 150 leader-follower dyads that when leaders' empowering behaviours did not match followers' empowerment expectations, followers perceived them as laissez-faire, which lowered their evaluations of leader effectiveness. Humborstad and Kuvaas (2013) examined mutuality in leaders' and subordinates' empowerment expectations and its relationship with role ambiguity and intrinsic motivation. Norris, Ghahremani and Lemoine (2021) found a three-way interaction between delegation, perceived leader competence and subordinate gender in predicting perceptions of laissez-faire leadership and dysfunctional resistance. Skogstad and colleagues (2007) surveyed 2,273 Norwegian employees and found laissez-faire leadership correlated with role conflict, role ambiguity and conflicts with coworkers, which mediated its effects on bullying and psychological distress. Skogstad and colleagues (2014) found in a three-wave prospective study of 1,771 employees that laissez-faire leadership was positively related to role ambiguity at all three points, while initiating structure and consideration were not related to role ambiguity. Hinkin and Schriesheim (2008) found that leader reward omission and punishment omission related to follower satisfaction with the leader, perceived leader effectiveness, how clear followers were about their roles, and supervisor-rated performance. Gonzalez-Mulé and colleagues (2016) studied 110 teams in a defence manufacturing firm and found that highly autonomous teams receiving a high degree of performance feedback outperformed other teams through greater understanding of the organisation's goals, while highly autonomous teams receiving little feedback performed worst. Langfred (2004) found that high trust could make members of self-managing teams reluctant to monitor one another, and that low monitoring combined with high individual autonomy could harm team performance. Hu and colleagues (2025) found in a three-wave study of 74 teams that differentiated empowering leadership reduced team information sharing and, through it, team task performance.
Being trusted, and what it asks of you
Most research on trust at work asks how much people trust their manager. A smaller and very interesting body of work asks the other question: how much do people feel their manager trusts them? It turns out to matter a great deal, and in ways that are not quite what you would expect.
A 2003 study found that people's sense of being trusted by their leader predicted their performance, their willingness to go beyond the job and their satisfaction, and predicted all three more strongly than how trustworthy they thought their leader was. A longitudinal study of 88 retail stores found that when employees collectively felt trusted by management, a norm of responsibility grew up among them, and sales and customer service improved; the norm of responsibility carried the whole effect on sales. Being trusted made people act like owners.
Which kind of trust matters? A study of 497 teachers separated feeling relied upon from being confided in. Being relied upon raised people's sense of their own worth at work and, through it, their performance. Being confided in did not. The trust that counts is the trust that hands you something that matters. Studies of pairs of managers and their direct reports found that trust in each direction had its own effects, and that performance and helpfulness were highest when trust ran both ways at once.
Then the counterweight, and it is an important one for anyone who has been the most trusted person in the building. A study of bus drivers in London found that feeling trusted was a double-edged sword. It filled people with pride, which was good for them. It also increased their sense of workload and their worry about keeping up their reputation, and those burdens were exhausting. A 2026 study added that people who already felt secure in their value at work responded to being trusted with more engagement, and people who felt less secure responded with less engagement and more work spilling into home. Trust can be a gift and a weight at the same time, and it helps to say so.
Being trusted makes people act like owners. It also hands them an owner's worries, and a good leader notices both.
How is trust built between two people who start from caution? The research on this is encouraging, because it is mostly about behaviour rather than personality. A six-week field study of teams who alternately built a system and supervised another team's build found that when one side took a risk on the other, the other side saw them as more trustworthy, trusted them more and took more risks in return, a pattern that repeated through the whole project. A study of pairs found that perceived trustworthiness and cooperation spiral: each move by one person shapes the next move by the other, upwards or downwards. A 2018 set of studies found that trust trickles down: when managers were trusted by their own bosses, especially when they were involved in decisions, they behaved in more trusting ways towards their teams, and their teams performed better.
A classic paper on managers as the initiators of trust named the behaviours that start the spiral: being consistent, keeping words and deeds aligned, sharing control and delegating, communicating accurately and openly with explanations for decisions, and showing concern for the other person's interests. A later field study found that trustworthy behaviour of this kind built trust even in the middle of conflict. And a field study found the practical payoff: when people trusted management, they could stop spending attention on protecting themselves and focus on the work.
Put those together and something useful emerges for the seat you sit in. Trust is built by small, visible risks taken in both directions and honoured. Your manager takes a risk every time they let something go without checking. You take one every time you flag a problem early instead of hiding it. The spiral goes up when each of those risks is repaid.
Try this. Think of one thing your manager currently checks that they could reasonably stop checking. Ask to own it for a month, with a clear way for them to see the outcome without seeing every step. That is the small, bounded risk the reciprocal trust study found starts the upward spiral.
The research behind this section
Lester and Brower (2003) found that subordinates' felt trustworthiness, their perception of their leader's trust in them, was positively related to performance, organisational citizenship behaviour and job satisfaction, and was a more significant predictor of these outcomes than their perceptions of their leader's trustworthiness. Salamon and Robinson (2008) conducted a longitudinal study of 88 retail stores and found that employees' collective perceptions of being trusted by management increased responsibility norms and sales and customer service performance, with responsibility norms fully mediating the effect on sales. Lau, Lam and Wen (2014) studied 497 teachers and found that felt trust measured as reliance, but not as disclosure, enhanced organisation-based self-esteem, which mediated its effect on work performance. Brower and colleagues (2009) found unique relationships of trust in the manager and trust in the subordinate with performance, citizenship and intentions to quit, and that their interaction predicted individual-directed citizenship. Kim, Wang and Chen (2018) found in 247 subordinate-supervisor pairs a positive effect of perceived mutual trust on task performance and interpersonal facilitation after controlling for trust in the leader and felt trust. Skiba and Wildman (2019) found in 208 employees that trust in the supervisor and felt trust interacted to reduce turnover intention through reduced uncertainty, while felt trust increased engagement through felt obligation and autonomy. Baer and colleagues (2015) found in a field study of London bus drivers that feeling trusted increased pride, which benefited exhaustion and performance, and also increased perceived workload and reputation maintenance concerns, which burdened them. Kong and colleagues (2026) found across a multi-wave survey in the United States and a cross-lagged survey in China that employees with higher organisation-based self-esteem responded to felt trust with higher work engagement and better task performance, and those with lower self-esteem responded with lower engagement and more work-to-home conflict. Serva, Fuller and Mayer (2005) found in a six-week field study of 24 interacting teams that risk-taking actions by one team predicted the other team's trustworthiness perceptions and trust, which predicted its subsequent risk-taking in return, in a repeating reciprocal pattern. Ferrin, Bligh and Kohles (2008) found through an interdependence analysis that perceived trustworthiness and cooperation spiral within dyads. De Cremer and colleagues (2018) found across a field study and two experiments that perceptions of trustworthiness trickled down the hierarchy to enhance subordinate performance, especially when managers invited supervisors into decision-making, mediated by supervisors' more trusting behaviour. Whitener and colleagues (1998) set out a framework of managerial trustworthy behaviour: behavioural consistency, behavioural integrity, sharing and delegation of control, communication, and demonstration of concern. Korsgaard, Brodt and Whitener (2002) found that managerial trustworthy behaviour built subordinates' trust in the face of conflict. Mayer and Gavin (2005) found that trust in management related to employees' ability to focus attention on value-producing activities, which related to performance and citizenship. Kramer (1999) reviewed research on trust and distrust in organisations. De Jong, Dirks and Gillespie (2016) meta-analysed 112 studies of 7,763 teams and found that intrateam trust was positively related to team performance (ρ = .30), more so under high task interdependence.
Working well for someone who checks everything
So what do you do on the Tuesday when the fourth "where are we on this?" arrives? The research offers no trick, and it points consistently in one direction: the fastest way to reduce checking is to make it unnecessary, predictable and useful, in that order.
Start by assuming the checking is about something. The teachers' study found that pressure from above makes people controlling below. The 1958 study found that watching creates its own doubt. The micromanagement studies found low trust and a thin working relationship among its roots. Very often the person checking on you is carrying a worry you cannot see: a board member who is asking them, a previous assistant who dropped something, a project whose failure would land on them. The micromanager I worked for taught me not to sweat the small stuff, even while making me sweat over it, and part of that lesson was seeing that the questions were rarely about me.
Get ahead of the question. The diary studies found that unpredictable monitoring hurt felt trust more than regular monitoring. The practical answer is to become the predictable source of information yourself, so the checking has less reason to happen at random. A short update at the same time each day or each week, in the same shape, answers the question before it is asked.
Make the update a conversation, not a report. Interactional monitoring, checking in through conversation, went with trust in a way that watching from a distance did not. A brief, regular exchange where you say what is done, what is next and what you need will usually replace a scatter of messages. Offer the conversation rather than waiting to be summoned to it.
Propose the rhythm explicitly. The research on empowerment expectations found that mismatched expectations, in either direction, are where things go wrong. So say out loud what you are proposing: which decisions you will make and tell them about afterwards, which you will bring to them with a recommendation, and which you will leave entirely to them. Ask for a review date, because micromanagement is defined in part by being sustained, and a date gives both of you a way out.
Build the record in small, visible steps. The reciprocal trust study found that trust grew when one side took a risk and the other honoured it. Your version is to ask for one bounded piece of ownership at a time and to make the outcome visible without making every step visible. When it goes well, say so briefly; when something slips, say that first and fast. People forgive honest early warnings much more readily than late surprises, and the record you build is the thing that eventually lets them stop checking.
There are limits. Some managers check because of something in them rather than something in the situation, and no rhythm will fully cure that. And there are moments, a new executive, a crisis, an audit, where close attention is right and a good Chief of Staff will welcome it. The aim is proportion rather than freedom from all oversight.
Try this. Write a three-line proposal and send it this week: "To save you chasing, I will send a two-minute update every [day] at [time]: what is done, what is next, anything I need from you. I will decide [these things] and tell you afterwards, and bring [these] to you first. Shall we try it for a month and see?"
The research behind this section
Pelletier, Séguin-Lévesque and Legault (2002) found that pressure from above made teachers more controlling towards their students. Strickland (1958) found that close surveillance led supervisors to distrust the worker they surveilled. Deen and colleagues (2025) found that micromanagement related to low trust and a weaker leader-member exchange. Zheng and colleagues (2023) found that monitoring variability intensified the negative effect of daily monitoring on felt trust in a newly remote context. Liao and Chun (2016) distinguished observational from interactional monitoring and found that they related differently to trust and distrust in the supervisor. Wong and Giessner (2018) found that empowering behaviour that did not match followers' expectations was perceived as laissez-faire. Humborstad and Kuvaas (2013) examined the mutuality of leader and subordinate empowerment expectations. Serva, Fuller and Mayer (2005) found that risk-taking actions by one party predicted the other's trust and subsequent risk-taking in a reciprocal pattern. Gonzalez-Mulé and colleagues (2016) found that autonomy combined with performance feedback produced the best team performance. Thiel and colleagues (2023) found that developmental feedback buffered the relational costs of monitoring. Skogstad and colleagues (2007) found that laissez-faire leadership was a destructive rather than a neutral form of leadership.
When you are the one holding the reins
Many people reading this also manage someone: a junior assistant, a team coordinator, a project group, or the whole support function for a leadership team. The research asks something harder of you than of the person being managed, because you are the one who can start the spiral.
Delegate something that matters, and say why. Delegation is one of the clearest signals of trust a manager can send. A study in China found that delegation went with better work outcomes through people's sense of being insiders and their self-esteem at work, and an earlier study found that delegation and a high-quality working relationship went together. The teachers' study on felt trust suggests the sort of delegation that counts: being relied on for something real, not confided in about something trivial.
Match the structure to the person and the moment. The directive teams did better at first and the empowered teams overtook them; the least experienced salespeople gained most from empowerment; people forgave extra consideration and resented extra structure. Put together, that suggests giving more structure early in something new and taking it away deliberately as the person grows into the work. Tell them that is what you are doing, so the change reads as a promotion and not as a withdrawal.
Pair freedom with feedback. The channelled autonomy finding is the one I would pin up. People who were given freedom and regular feedback performed best; people given freedom with little feedback performed worst. The feedback does not need to be elaborate. It needs to exist, and to come back to the person as help rather than disappear upwards as evidence.
Make it safe to bring you bad news. A study of 51 work teams in a manufacturing company found that psychological safety, a shared belief that it is safe to take interpersonal risks, went with learning behaviour, which in turn went with performance. A meta-analysis of 136 samples, more than 22,000 people and nearly 5,000 groups, confirmed that psychological safety predicted performance and going beyond the job even after accounting for related ideas such as good relationships with the leader. The single most useful thing a manager can do to reduce the need for checking is to make early warnings welcome, because then the problems come to you before you have to go looking for them.
Trust evenly, and watch your own pressure. The finding that uneven empowerment hurt team information sharing is a caution against having favourites in the matter of freedom. The teachers' study is a caution to look at the pressure you are under, because it is likely to show up in how closely you manage.
And the encouraging part: this can be learned. Managers trained in autonomy support changed their style within five weeks, and their people noticed. The leader I worked for who asked "how can I support you?" was not born knowing to ask it.
Try this. Before you next hand something over, fill in four lines: what good looks like, what they decide alone, when you will look at it together, and when you will stop looking. Share the four lines with them. Then keep to the last one.
The research behind this section
Chen and Aryee (2007) examined delegation in China and found that it related to employee work outcomes through organisation-based self-esteem and perceived insider status. Schriesheim, Neider and Scandura (1998) examined delegation and leader-member exchange, its main effects, moderators and measurement. Lau, Lam and Wen (2014) found that felt trust in the form of reliance, but not disclosure, raised performance through organisation-based self-esteem. Lorinkova, Pearsall and Sims (2013) found that directive leadership produced better early team performance and empowering leadership greater improvement over time. Ahearne, Mathieu and Rapp (2005) found that empowering leadership benefited less experienced and less knowledgeable sales representatives most. Lambert and colleagues (2012) found that excess initiating structure, but not excess consideration, was associated with unfavourable attitudes. Gonzalez-Mulé and colleagues (2016) found that autonomy with feedback produced the highest team performance and autonomy without feedback the lowest. Edmondson (1999) studied 51 work teams in a manufacturing company and found that team psychological safety was associated with learning behaviour, which mediated between psychological safety and team performance. Frazier and colleagues (2017) meta-analysed 136 independent samples representing over 22,000 individuals and nearly 5,000 groups and found that psychological safety influenced task performance and citizenship behaviour over and above related concepts such as positive leader relations and work engagement. Hu and colleagues (2025) found that differentiated empowering leadership reduced team information sharing and performance. Hardré and Reeve (2009) found that training changed managers' motivating style within five weeks.
The check-in check
Everything in this post comes down to one recurring moment: somebody is checking on your work more than it needs, and you want to change the pattern without damaging the relationship. So here is a tool for that moment.
Tell it how often you are being checked, what the checking looks like, how new the work is and what the stakes are, whether you have ever agreed how updates should work, and what pressure the other person might be under. It will give you a reading of what the checking is most likely about, the steps the research points to, what to watch for afterwards, and a reporting rhythm you can copy and propose.
Where to start
If you take one thing from all of this, make it the rhythm. Most of the checking that wears people down is unpredictable, unexplained and aimed at the wrong things, and a short, regular, agreed update fixes all three at once. It gives the other person the information they were looking for, in a form that feels like conversation rather than surveillance, and it builds the record that eventually lets them stop.
After that, there are three moves worth making.
When you are checked on, ask what the check is for and offer to answer it before it is asked. When you are trusted with something, ask what good looks like and when you should show your progress, because freedom with feedback is what performs. And when you manage others, hand over something that matters, match the structure to the moment, and tell them when you will stop looking.
I have put the three pages I would use into a download: a reporting rhythm to agree with your manager, a record of commitments that builds trust in small visible steps, and a handing-over sheet for when you are the one delegating.
Trust turns out to be less a feeling than a practice, built out of small risks taken and honoured, and it is one that people in your role are unusually well placed to lead. You sit where the information flows, you see where the worry is, and you can make the checking unnecessary long before anyone thinks to stop. That is a quiet kind of leadership, and a very good one.
The part of this that shows up in how you are seen at work sits inside what the Connected Leader Assessment calls Influence and Stakeholder Trust: whether the people you work with rely on you without needing to check, how you build confidence with a new executive, and whether your record makes the case for your judgement before you have to make it yourself. It is the dimension that decides how much room you are given to do your best work.
The assessment is free, takes about twelve minutes and gives you a read across all five dimensions. If you want to go further on this one, the Skill Sprints are short, focused programmes on a single dimension, and the Irreplaceable Audit is the place to look at which parts of your work your executive would never think to check, because those are the proof of trust worth building on.
Stay savvy.
Meg ✌️
Free tools
The check-in check, the interactive tool in this post, free
The Trust Pages, the download in this post, free
The Connected Leader Assessment, twelve minutes, free, with a read on Influence and Stakeholder Trust, the dimension this post sits in
The Irreplaceable Audit, forty-eight activities across four kinds of work, free, for spotting the parts of your work nobody needs to check
P.S. Want to Go Deeper on Micromanagement and Trust?
If this resonated and you want to go further, here is the evidence the piece is built on. These are the papers that shaped my thinking on what close watching costs, what autonomy gives, and how trust is built in small steps, and they are well worth your time.
Every DOI below has been checked. Where I could not confirm a figure from the paper itself, I have left the figure out rather than guess.
📃 Judge, Piccolo and Ilies (2004) Meta-analysis. Consideration (.48) and initiating structure (.29) both related to leadership outcomes. https://doi.org/10.1037/0021-9010.89.1.36
📃 DeRue and colleagues (2011) Leader traits and behaviours together explained at least 31% of leadership effectiveness. https://doi.org/10.1111/j.1744-6570.2010.01201.x
📃 Judge and Piccolo (2004) Meta-analysis of transformational and transactional leadership and their relative validity. https://doi.org/10.1037/0021-9010.89.5.755
📃 Kerr and Jermier (1978) The idea that features of the task, people and organisation can substitute for leadership. https://doi.org/10.1016/0030-5073(78)90023-5
📃 Podsakoff, MacKenzie and Bommer (1996) Meta-analysis of the substitutes for leadership and their effects on attitudes and performance. https://doi.org/10.1037/0021-9010.81.4.380
📃 Deen and colleagues (2025) Micromanagement defined and measured in 1,723 people: controlling, close and detailed, and excessive, sustained and unnecessary. https://doi.org/10.1177/01492063251378092
📃 Liao and Chun (2016) 388 pairs. Watching from a distance and checking in through conversation related differently to trust. https://doi.org/10.1002/job.2035
📃 Zheng and colleagues (2023) Two diary studies. More monitoring on a day meant feeling less trusted and more exhausted that day. https://doi.org/10.1002/job.2699
📃 Thiel and colleagues (2023) Monitoring eroded the working relationship; giving the data back as developmental feedback buffered it. https://doi.org/10.1002/job.2748
📃 Thiel, Prince and Sahatjian (2022) Electronic monitoring undermined the trust-building of otherwise ethical leaders. https://doi.org/10.1111/1748-8583.12462
📃 Ravid and colleagues (2023) 94 samples, 23,461 workers. Monitoring did not improve performance, raised stress, and transparency helped attitudes. https://doi.org/10.1111/peps.12514
📃 Ravid and colleagues (2020) A review and typology of electronic performance monitoring, and why its findings conflict. https://doi.org/10.1177/0149206319869435
📃 Siegel, König and Lazar (2022) Meta-analysis. Small costs to satisfaction and stress, and no link with performance. https://doi.org/10.1016/j.chbr.2022.100227
📃 König (2025) Review of electronic monitoring at work: neutral on performance, small costs to strain and attitudes. https://doi.org/10.1146/annurev-orgpsych-110622-060758
📃 Aiello and Kolb (1995) Monitoring made skilled people faster and less skilled people slower; the unmonitored felt least stressed. https://doi.org/10.1037/0021-9010.80.3.339
📃 Kolb and Aiello (1996) Whether monitoring stressed people depended on whether they felt in control of their own outcomes. https://doi.org/10.1016/0747-5632(96)00016-7
📃 Miao and colleagues (2025) 3,744 remote doctors. Controlling monitoring raised burnout; developmental monitoring lowered it. https://doi.org/10.1080/14719037.2024.2396080
📃 Strickland (1958) Supervisors came to trust the worker they watched closely less than an identical one they did not. https://doi.org/10.1111/j.1467-6494.1958.tb01580.x
📃 Bernstein (2012) The transparency paradox. A modest zone of privacy on a factory line raised performance. https://doi.org/10.1177/0001839212453028
📃 Falk and Kosfeld (2006) The hidden costs of control. Most people who were controlled reduced effort, reading it as distrust. https://doi.org/10.1257/aer.96.5.1611
📃 Ziegelmeyer, Schmelz and Ploner (2012) Four repetitions, 476 people. The hidden costs were real but usually smaller than the benefit of control. https://doi.org/10.1007/s10683-011-9302-8
📃 Riener and Wiederhold (2016) People who had first worked together as a team reacted even more strongly to being controlled. https://doi.org/10.1016/j.jebo.2015.12.008
📃 Sliwka (2007) A model in which choosing control signals what you expect of people, and they respond to the signal. https://doi.org/10.1257/aer.97.3.999
📃 Frey and Jegen (2001) A survey of the evidence on when outside intervention crowds out motivation. https://doi.org/10.1111/1467-6419.00150
📃 Bengtsson and Engström (2014) A randomised policy experiment in which closer monitoring improved outreach and cut irregularities. https://doi.org/10.1111/ecoj.12049
📃 Deci, Olafsen and Ryan (2017) Self-determination theory in work organisations: the state of the science. https://doi.org/10.1146/annurev-orgpsych-032516-113108
📃 Gagné and Deci (2005) Self-determination theory as a theory of work motivation, autonomous and controlled. https://doi.org/10.1002/job.322
📃 Van den Broeck and colleagues (2016) 99 studies. Meeting the needs for autonomy, competence and relatedness at work, and what follows. https://doi.org/10.1177/0149206316632058
📃 Slemp and colleagues (2018) 72 studies, 32,870 people. Leader autonomy support went with self-motivation, well-being and less distress. https://doi.org/10.1007/s11031-018-9698-y
📃 Slemp and colleagues (2024) 881 samples, 443,556 people. Supporting basic needs related strongly to well-being and modestly to performance. https://doi.org/10.1037/pspi0000459
📃 Slemp, Lee and Mossman (2021) Ten field interventions to make managers more need-supportive; most helped, more so for managers than teams. https://doi.org/10.1111/joop.12338
📃 Baard, Deci and Ryan (2004) Two workplaces. Managers' autonomy support met needs, which predicted performance ratings and adjustment. https://doi.org/10.1111/j.1559-1816.2004.tb02690.x
📃 Humphrey, Nahrgang and Morgeson (2007) Meta-analysis of work design features, including autonomy, and what they predict. https://doi.org/10.1037/0021-9010.92.5.1332
📃 Spector (1986) Meta-analysis. Feeling in control at work went with satisfaction, motivation, performance and less strain. https://doi.org/10.1177/001872678603901104
📃 Deci, Connell and Ryan (1989) Managers' support for self-determination and employees' satisfaction and trust in a large corporation. https://doi.org/10.1037/0021-9010.74.4.580
📃 Hardré and Reeve (2009) 25 managers trained in autonomy support; five weeks later their 169 employees were more engaged. https://doi.org/10.1111/j.1468-2419.2009.00325.x
📃 Pelletier, Séguin-Lévesque and Legault (2002) Teachers under pressure from above became more controlling with their students. https://doi.org/10.1037/0022-0663.94.1.186
📃 Lee, Willis and Tian (2018) 105 samples. Empowering leadership related to performance, citizenship and creativity through trust and empowerment. https://doi.org/10.1002/job.2220
📃 Kim, Beehr and Prewett (2018) 55 samples. Empowering leaders were liked (.59) more reliably than they raised performance (.31). https://doi.org/10.1177/1548051817750538
📃 Cheong and colleagues (2019) A review of why empowering leadership's results are mixed, and how to study it better. https://doi.org/10.1016/j.leaqua.2018.08.005
📃 Cheong and colleagues (2016) Empowering leadership enabled people through confidence and burdened them through tension. https://doi.org/10.1016/j.leaqua.2016.01.006
📃 Sharma and Kirkman (2015) A review setting out the unintended, negative side of empowering leadership. https://doi.org/10.1177/1059601115574906
📃 Ahearne, Mathieu and Rapp (2005) Empowering leadership helped the least experienced and least knowledgeable salespeople most. https://doi.org/10.1037/0021-9010.90.5.945
📃 Wong and Giessner (2018) 150 pairs. Empowerment that missed the follower's expectations was seen as laissez-faire. https://doi.org/10.1177/0149206315574597
📃 Humborstad and Kuvaas (2013) Whether leaders and employees expected the same amount of empowerment, and why it matters. https://doi.org/10.1016/j.leaqua.2013.01.003
📃 Lorinkova, Pearsall and Sims (2013) 60 teams. Directed teams started better; empowered teams learned faster and overtook them. https://doi.org/10.5465/amj.2011.0132
📃 Martin, Liao and Campbell (2013) A field experiment. Both styles raised proficiency; only empowering leadership raised proactivity. https://doi.org/10.5465/amj.2011.0113
📃 Sanchez-Manzanares and colleagues (2020) 67 teams in a simulation. Directed teams did better both before and after a sudden change. https://doi.org/10.1177/1059601120958838
📃 Somech (2005) 140 school teams. Directive leadership bought commitment and performance; participative, innovation. https://doi.org/10.1177/0013161x05279448
📃 Burke and colleagues (2006) Team meta-analysis. Task and person focus both mattered; empowerment explained nearly 30% of team learning. https://doi.org/10.1016/j.leaqua.2006.02.007
📃 Langfred (2004) In self-managing teams, high trust and high autonomy could mean too little monitoring and worse results. https://doi.org/10.2307/20159588
📃 Gonzalez-Mulé and colleagues (2016) 110 teams. Autonomy with feedback performed best; autonomy without feedback performed worst. https://doi.org/10.1177/0149206314535443
📃 Seibert, Wang and Courtright (2011) Meta-analysis of what builds psychological and team empowerment and what it leads to. https://doi.org/10.1037/a0022676
📃 Skogstad and colleagues (2007) 2,273 employees. Laissez-faire leadership went with role stress, bullying and distress. https://doi.org/10.1037/1076-8998.12.1.80
📃 Skogstad and colleagues (2014) Three waves, 1,771 employees. Laissez-faire leadership predicted role ambiguity; structure and consideration did not. https://doi.org/10.1080/02678373.2014.957362
📃 Hinkin and Schriesheim (2008) Leaders who did not respond to good or poor performance were rated less effective. https://doi.org/10.1037/a0012875
📃 Norris, Ghahremani and Lemoine (2021) Delegation could be read as laissez-faire leadership, depending on how competent the leader seemed. https://doi.org/10.1177/1548051821997407
📃 Thompson and Vecchio (2009) A test of three versions of situational leadership theory found little support for its prescriptions. https://doi.org/10.1016/j.leaqua.2009.06.014
📃 Salamon and Robinson (2008) 88 retail stores. Feeling trusted by management built responsibility norms and raised sales. https://doi.org/10.1037/0021-9010.93.3.593
📃 Lau, Lam and Wen (2014) 497 teachers. Being relied on raised performance through self-esteem at work; being confided in did not. https://doi.org/10.1002/job.1861
📃 Baer and colleagues (2015) London bus drivers. Feeling trusted brought pride, and also workload and reputation worries. https://doi.org/10.5465/amj.2014.0246
📃 Skiba and Wildman (2019) 208 employees. Trusting the boss reduced uncertainty; feeling trusted raised engagement. https://doi.org/10.1007/s10869-018-9537-x
📃 Kim, Wang and Chen (2018) 247 pairs. Performance was highest when trust ran both ways. https://doi.org/10.1007/s10551-016-3093-y
📃 Kong and colleagues (2026) Felt trust raised engagement for people secure in their worth, and lowered it for others. https://doi.org/10.1111/joms.13210
📃 Brower and colleagues (2009) Trust in the manager and trust in the subordinate each had their own effects on performance. https://doi.org/10.1177/0149206307312511
📃 Serva, Fuller and Mayer (2005) 24 teams over six weeks. A risk taken by one side built the other's trust, and it came back. https://doi.org/10.1002/job.331
📃 Ferrin, Bligh and Kohles (2008) Perceived trustworthiness and cooperation spiral up or down between two people. https://doi.org/10.1016/j.obhdp.2008.02.012
📃 Mayer and Gavin (2005) Trust in management let employees focus on the work, which related to performance. https://doi.org/10.5465/amj.2005.18803928
📃 Korsgaard, Brodt and Whitener (2002) Managers' trustworthy behaviour built trust even in the middle of conflict. https://doi.org/10.1037/0021-9010.87.2.312
📃 Whitener and colleagues (1998) Managers as initiators of trust: the behaviours that start it. https://doi.org/10.5465/amr.1998.926624
📃 Lester and Brower (2003) Feeling trusted predicted performance and satisfaction more strongly than trusting the leader did. https://doi.org/10.1177/107179190301000203
📃 Kramer (1999) Review of trust and distrust in organisations. https://doi.org/10.1146/annurev.psych.50.1.569
📃 De Cremer and colleagues (2018) Trust trickled down: managers who were trusted behaved more trustingly, and their teams performed better. https://doi.org/10.1037/apl0000339
📃 Edmondson (1999) 51 teams. Psychological safety went with learning behaviour, which went with performance. https://doi.org/10.2307/2666999
📃 Frazier and colleagues (2017) 136 samples, 22,000 people. Psychological safety predicted performance beyond related ideas. https://doi.org/10.1111/peps.12183
📃 Chen and Aryee (2007) In China, delegation related to better outcomes through feeling like an insider. https://doi.org/10.5465/amj.2007.24162389
📃 Schriesheim, Neider and Scandura (1998) Delegation and the quality of the working relationship between leader and employee. https://doi.org/10.2307/256909
📃 De Jong, Dirks and Gillespie (2016) 112 studies, 7,763 teams. Trust within teams related to performance (.30). https://doi.org/10.1037/apl0000110
📃 Chen and colleagues (2024) An inverted U between empowering leadership and team performance, through reflection. https://doi.org/10.1177/10464964241231570
📃 Hu and colleagues (2025) 74 teams. Empowering some members much more than others cut information sharing and performance. https://doi.org/10.1177/00187267241303265
📃 Lambert and colleagues (2012) Too little structure or consideration hurt; too much consideration was fine, too much structure was not. https://doi.org/10.1037/a0028970
📃 Schweitzer, Ho and Zhang (2018) People complied when watched and exploited when not, and the watchers mistook compliance for trustworthiness. https://doi.org/10.1287/mnsc.2016.2586
📃 Lee and colleagues (2017) 137 pairs. Performance rose with empowering leadership and then fell, depending on learning orientation. https://doi.org/10.1177/1059601116646474
P.P.S. Why Do I Even Have the Nerve to Write This?
Fair question.
I am someone who has spent years working alongside remarkable executives, navigating chaos, translating vision into structure and figuring things out, sometimes beautifully and sometimes the very hard way. I have had the privilege of being the steady right hand to leaders who move fast, take risks and expect a lot. That environment has shaped me more than any classroom ever could.
I also happen to have an MBA, and I am studying psychology because people fascinate me. How we work. Why we disconnect. What actually holds us together when the pressure rises.
But really, none of that is the point.
I am here because I have lived the erosion and rebuilt from it. I have seen the cost of carrying too much and the relief that arrives when you finally stop disappearing inside your own competence.
So I share what I have learned in case it helps someone else. Take what is useful. Leave the rest. And remember this one truth that professionals like you often forget:
You are probably doing far better than you give yourself credit for.
Comments