Status: Companion paper
Relationship: Companion to The Science of Management Failures
Management is not only performed. It must also be reproduced.
Organizations cannot assume that executive judgment will appear whenever an executive position becomes vacant. They must transmit knowledge of priorities, relationships, exceptions, institutional history, and political context from one generation of role-holders to the next. Formal authority can be conferred at once. The judgment required to exercise it usually cannot.
This paper examines management reproduction as an institutional process. Its central mechanism is apprenticeship: a prospective manager participates in the work of an experienced manager, first as an observer and adviser, then as a representative and delegated decision-maker, and eventually as an independent occupant of a managerial role. Executive assistants provide one vivid example because their visible administrative work may coexist with a largely invisible process of executive formation. Similar arrangements appear in law firms, hospitals, universities, militaries, governments, religious bodies, and family enterprises.
The paper distinguishes authority from judgment, describes progressive delegation and executive recursion, and identifies characteristic failures of managerial reproduction. It concludes with implications for Pancakes: systems that record institutional activity should distinguish operational, managerial, and executive observations; preserve provenance and uncertainty; and avoid inferring authority, competence, or succession from proximity alone.
The Science of Management Failures asks a fundamental question:
How does management fail?
This paper asks the complementary question:
How is management reproduced?
The first question concerns pathological enactments of a managerial role. A manager may fail to coordinate work, allocate resources, communicate decisions, resolve conflicts, or maintain the conditions under which others can succeed. The second question concerns continuity. How does an institution ensure that someone will be able to perform those functions after the current manager leaves?
Job descriptions provide only a partial answer. They can identify responsibilities, reporting relationships, and formal powers. They cannot fully communicate which issues require immediate intervention, which stakeholders must be consulted, which apparent rules admit exceptions, why an old compromise still matters, or how a decision will be interpreted by people with long organizational memories. These are elements of practical judgment. They are learned through participation in institutional life.
Management reproduction is therefore more than recruitment and more than promotion. It is the process by which an institution recreates the capacity to occupy a managerial role. That capacity includes technical knowledge, but it also includes situated judgment: the ability to perceive what kind of situation is occurring, determine what matters within it, recognize who is entitled to act, and choose an action that the institution can sustain.
This process is easiest to see in professions that explicitly name apprenticeship. A junior physician observes an experienced physician, discusses cases, undertakes supervised work, and acquires greater independence. A junior lawyer moves from research to client contact to responsibility for a matter. A military officer is progressively entrusted with command. Management often follows the same pattern without describing itself in those terms.
One revealing case is the executive assistant who accompanies a senior executive into meetings, prepares decisions, coordinates follow-up, communicates on the executive’s behalf, and later becomes the leader of a division. To people who saw mainly calendar management and administrative support, the promotion may look arbitrary or incompatible with the assistant’s previous role. To people who saw a long apprenticeship in executive work, it may look like the completion of an established succession process.
The point is not that executive assistants are secretly executives, nor that proximity to power proves preparation for power. The point is that visible tasks do not always reveal the institutional function of a role. An administrative role can also be an observational post, a relationship hub, a delegated agency, or a site of formation. Understanding succession requires looking not only at what work is produced, but at what capability is being cultivated.
Institutions produce outputs, but durable institutions also reproduce the roles that make those outputs possible.
Schools educate students while also developing teachers, principals, and administrators. Hospitals treat patients while training physicians, nurses, and clinical leaders. Courts decide cases while initiating lawyers into legal practice and preparing judges for more complex jurisdictions. Partnerships serve clients while cultivating future partners. Governments administer public programs while developing civil servants and political executives.
This suggests a general proposition:
Institutions reproduce roles, not merely outputs.
A role is not simply a collection of tasks assigned to an individual. It is a durable institutional position with expectations, relationships, powers, obligations, and characteristic forms of judgment. A functioning institution must repeatedly place new people into such positions without requiring them to invent the role from nothing.
Succession is therefore an institutional operation. It has inputs, mechanisms, dependencies, and failure modes. It may be deliberate or accidental, public or concealed, inclusive or hereditary, robust or fragile. Even when an organization claims to select the best candidate in an open competition, it depends upon some prior process that made candidates capable of understanding the work. Selection chooses among capacities that development has already produced.
The reproduction of a role has at least four dimensions:
These dimensions can separate. A successor may receive formal authority without the necessary judgment. An adviser may possess excellent judgment but no mandate to act. A chosen heir may be accepted by a narrow executive circle but regarded as illegitimate by the wider organization. A technically capable outsider may lack the historical relationships required to make decisions effective.
Reproduction is complete only when the new occupant can perform the role and the surrounding institution can recognize, constrain, and cooperate with that performance.
Formal authority can be assigned overnight. Judgment cannot.
A board can appoint a chief executive by resolution. A chief executive can name a division president in an announcement. Access rights, budgets, reporting lines, and signing powers can be changed within hours. None of those actions ensures that the appointee knows how to use the authority well.
Managerial judgment depends heavily upon tacit knowledge. Some of that knowledge can be documented, but much of it gains meaning only in context. It includes:
A procedure can specify who approves an expenditure. It is less able to explain when a modest expenditure expresses a major commitment, when refusing it will damage an essential relationship, or when approval would create a precedent that the organization cannot sustain. A briefing book can list stakeholders. It cannot reproduce years of interaction with them.
This is the apprenticeship problem: a role requires knowledge that cannot be transferred adequately through description, yet the institution must somehow transfer it.
The usual solution is participation. The learner encounters concrete cases alongside an experienced role-holder. They observe not only the final decision but also what the executive notices, which questions the executive asks, whose interpretation is trusted, what remains deliberately undecided, and how the decision is communicated. They then take on parts of the reasoning and action under conditions in which errors can still be detected and corrected.
Apprenticeship is not merely imitation. Mature formation should expose reasons to scrutiny rather than reproduce habits mechanically. The apprentice must eventually learn when the predecessor’s usual response does not fit. The objective is not to create a copy of the current executive. It is to develop a new person capable of carrying the institutional role through circumstances the predecessor has never encountered.
Executive assistance is commonly understood through its visible outputs: schedules, briefings, correspondence, travel, meeting preparation, and follow-up. Those outputs are real and essential. In some organizations, however, they describe only one layer of the role.
An assistant who travels with a chief executive, attends sensitive meetings, reviews incoming information, drafts responses, tracks commitments, and coordinates decisions occupies an unusual institutional location. The assistant may see the organization longitudinally and horizontally. Functional leaders see deeply into their own domains; the assistant may see how questions move between domains and how the executive integrates them.
Over time, such an assistant may function as several things at once:
The apprenticeship can remain invisible because observers classify the role by its most legible tasks. They see scheduling rather than exposure to institutional reasoning. They see note-taking rather than access to decision formation. They see physical proximity and may interpret it as personal favoritism; insiders may see repeated testing, increasing responsibility, and executive development.
Neither interpretation should be accepted automatically. Proximity can enable apprenticeship, but it can also enable patronage. Administrative excellence can coexist with executive potential, but one does not prove the other. A promotion that makes sense within a private succession process may still be unfair, opaque, or poorly governed from the perspective of the wider institution.
The analytical lesson is narrower and more useful:
Apparent administrative support may conceal a second institutional function: the reproduction of executive capability.
This pattern is not unique to executive assistants. Chiefs of staff, deputies, special advisers, residents, clerks, aides-de-camp, vice-principals, associate deans, junior partners, and project leads may all occupy roles that combine present service with preparation for future authority.
The conceptual heart of management reproduction is the distinction between authority and judgment.
Authority answers:
Who may decide?
Judgment answers:
Who knows how to decide?
Authority is relational and institutional. It exists when others recognize that a person is entitled to commit resources, direct action, resolve a dispute, or speak for the organization within some domain. Judgment is a capability. It consists in recognizing relevant features of a situation and choosing appropriately among possible actions.
They are independent organizational resources:
| Authority | Judgment | Characteristic situation |
|---|---|---|
| Low | Low | New employee or novice entering the domain |
| High | Low | Premature promotion or unsupported appointment |
| Low | High | Trusted adviser, experienced specialist, or former role-holder |
| High | High | Mature executive operating within a legitimate mandate |
This table describes conditions, not permanent kinds of people. Both authority and judgment are domain-specific. A mature finance executive may have high judgment in capital allocation and low judgment in clinical operations. A chief executive may have broad authority while properly deferring to a specialist whose expertise is deeper in a particular case.
Delegation can also mean several different things:
These forms should not be collapsed. Someone who prepares a recommendation may exercise substantial judgment without holding final authority. Someone authorized to sign may be executing a decision they did not make. Someone who represents an executive may have discretion in one setting and a strict script in another.
Healthy apprenticeship gradually aligns authority with demonstrated judgment. Too much authority too early creates preventable risk. Too much judgment without meaningful authority creates frustration, shadow governance, and dependence upon informal influence. When formal authority remains with one person while real decisions migrate to an unaccountable adviser, the institution loses the ability to see where power resides.
Authority must also remain connected to responsibility and accountability. Apprenticeship is not a justification for invisible power. If a person is making consequential decisions, affected participants should be able to determine the person’s mandate, the boundaries of delegation, the review mechanism, and who remains answerable for the result.
Shadowing is sometimes described as passive observation. In serious apprenticeship it is a structured mode of participation.
An apprentice may:
The object being learned is not simply the executive’s personal style. It is institutional reasoning: the interaction between a role, its environment, its obligations, and the particular human being occupying it.
Good shadowing makes this reasoning discussable. The experienced executive explains why a matter was escalated, why a seemingly attractive proposal was delayed, why a particular person was consulted, or why the normal rule was suspended. The apprentice proposes interpretations rather than merely absorbing conclusions. Differences become material for learning.
Bad shadowing produces mystique. The apprentice learns that senior people possess inscrutable instincts, that proximity is more important than reasons, and that institutional politics consists in anticipating the preferences of the powerful. This may reproduce compliance and courtcraft while failing to reproduce judgment.
The quality of an apprenticeship therefore depends upon reflective access, not mere exposure. Being present for a thousand decisions does not guarantee understanding them. The learner needs opportunities to formulate reasons, act on them, receive feedback, and revise their model of the institution.
Management capability is usually reproduced through a progressive migration of work and authority:
This sequence is not a universal ladder. Stages overlap, and different domains may progress at different speeds. An apprentice may independently manage routine operations while only observing a sensitive negotiation. A new executive may possess mature judgment in one field while requiring intensive support in another.
The value of progressive delegation lies in its feedback structure. It makes the apprentice’s judgment visible before the cost of error becomes intolerable. It lets the experienced executive distinguish a difference in style from a failure of reasoning. It also gives the surrounding organization time to build direct relationships with the future role-holder.
Each stage should contain four elements:
Delegation fails as formation when the learner receives tasks without context, responsibility without authority, authority without support, or criticism without access to the reasoning by which performance is judged.
Progressive delegation also requires eventual release. An executive who continually corrects, overrides, or privately redecides the successor’s choices preserves dependence instead of producing capability. The apprentice becomes an extension of the incumbent rather than an independent role-holder. Successful reproduction ends not in permanent supervision but in a new center of accountable judgment.
Managers coordinate work. Executives also reproduce managers.
This creates a recursive structure. A manager performs a role while preparing others to perform related roles. A division leader develops team leaders. A chief executive develops division leaders and possible successors. Those successors inherit not only operational responsibilities but an obligation to continue the process.
The organization thereby becomes capable of reproducing the conditions of its own coordination:
management → formation of managers → management
This recursion explains why leadership development cannot be separated entirely from ordinary executive work. Meetings, delegations, reviews, special assignments, and succession discussions do not merely allocate current labor. They shape who will possess organizational judgment in the future.
It also explains why some executives appear to spend surprisingly little time producing visible operational output. Their work may consist in choosing problems, constructing decision environments, assigning domains, interpreting ambiguous signals, maintaining relationships, and developing people whose decisions affect thousands of other actions. Operational workers encounter the outputs of this activity only indirectly.
Different institutional horizons follow. An operational participant sees the local task and the constraints immediately surrounding it. A manager sees dependencies across tasks and people. An executive sees commitments across divisions, time horizons, external relationships, and the continued existence of the institution. None of these perspectives is automatically correct, and higher position does not guarantee greater insight. They nonetheless make different facts salient.
The horizons are also interdependent. Executives who lose contact with operational reality may reproduce elegant but delusional models of the organization. Operational workers who cannot see institutional constraints may interpret necessary coordination as arbitrary interference. Healthy management creates channels through which each horizon can correct the others.
The reproduction of management can fail even when a successor is successfully appointed. These failures form a distinct class of management failure.
Close relationships create opportunities for observation and trust, but they also create opportunities for favoritism. If developmental access is distributed according to personal comfort, social similarity, or loyalty, the organization may later describe an unequal contest as meritocratic succession. The favored candidate appears uniquely prepared because others were denied the experiences through which preparation occurs.
An executive cohort may reproduce its own class, educational, racial, gender, professional, or cultural profile. Shared background is interpreted as executive presence; unfamiliarity is interpreted as lack of fit. The institution preserves continuity by narrowing the range of people entitled to embody authority.
Apprenticeship requires trust, but trust can become obedience to the incumbent. A successor may be selected for protecting the executive’s reputation, repeating preferred interpretations, or suppressing disagreement. This reproduces personal allegiance rather than institutional capability.
Tacit knowledge includes both wisdom and error. Apprentices learn what the institution notices, but also what it habitually ignores. A successful succession can therefore preserve discriminatory assumptions, failed strategies, misleading metrics, and normalized harms more effectively than a disruptive appointment would.
Some executives become irreplaceable by monopolizing relationships, context, and decision history. They may delegate action while retaining all interpretation. When they depart, documentation proves inadequate because no one else has practiced making the relevant judgments.
An apprentice formed entirely within the incumbent’s office may learn the executive’s model without learning the organization. Filtered briefings replace direct contact. The successor inherits confidence but not correction mechanisms.
High performance in a technical role does not by itself demonstrate the capacity to coordinate others, adjudicate competing goods, exercise authority, or steward an institution. Promotion may confer broad authority on someone whose judgment has never been tested outside a specialist domain.
An organization may search for someone with the predecessor’s charisma, habits, vocabulary, or biography. The successor is evaluated as an imitation rather than as a person capable of fulfilling the role under new conditions. This confuses one enactment of management with management itself.
An adviser may come to exercise substantial influence while remaining absent from formal governance. Decisions are attributed to the executive, while the adviser cannot be questioned, reviewed, or held responsible. Developmental delegation has become invisible rule.
A candidate may possess judgment and receive formal authority yet remain unable to lead because the process is regarded as secret, predetermined, or unfair. Institutional acceptance cannot always be commanded by appointment.
The incumbent delegates difficult work but withholds title, authority, compensation, or recognition. The apprentice performs executive labor without becoming an executive. Reproduction stalls because dependence benefits the current role-holder.
An organization may rely on external recruitment whenever a vacancy occurs. External candidates can bring essential knowledge and correction, but a system that develops no internal capability loses institutional memory and signals that contribution within the organization does not lead to meaningful stewardship.
The remedy is not a closed internal succession system. A healthy institution combines continuity with contestability. It develops multiple people, exposes developmental opportunities to review, documents mandates, invites outside perspectives, and tests inherited judgment against operational evidence and affected communities.
Pancakes is concerned with tools that support healthy institutions without replacing human judgment. The reproduction of management introduces several design requirements for representing institutional activity.
Pancakes should distinguish among at least three kinds of observation:
| Observation level | Typical subject | Characteristic horizon |
|---|---|---|
| Operational | Tasks, incidents, resources, immediate outcomes | What is happening in the work? |
| Managerial | Dependencies, allocation, coordination, team capacity | How is the work being organized? |
| Executive | Authority, commitments, institutional risk, succession | How is the institution being directed and continued? |
The categories describe the subject and horizon of an observation, not the rank of the observer. An operational worker may make an insightful executive observation. An executive may make an inaccurate operational one.
Seeing that an assistant attends meetings, drafts messages, or communicates decisions does not establish the scope of that person’s authority. Systems should not infer delegation from proximity or communication patterns alone.
Where authority matters, it should be represented explicitly:
An observation may provide evidence about the exercise of authority. It does not create authority.
Claims about executive work are often interpretive. “She scheduled every meeting,” “she was being trained as a successor,” and “she exercised delegated authority” are different claims supported by different evidence. Pancakes should preserve source, context, confidence, and the distinction between observed conduct and inferred institutional meaning.
Institutions benefit from knowing whether they are developing future managers. That does not justify recording every interaction or generating opaque scores of executive potential. A humane system can document developmental assignments, domains of delegated responsibility, mentorship commitments, learning goals, and review outcomes without claiming to calculate a person’s worth.
Delegation should not make responsibility disappear. Pancakes should make it possible to identify both the actor exercising delegated power and the role-holder accountable for establishing and supervising the mandate. This is particularly important when informal roles acquire consequential influence.
Succession records can expose whether developmental opportunities are broadly available or concentrated within an executive’s trusted circle. Appropriate governance may include:
Transparency need not mean publishing every confidential deliberation. It means making the structure of authority, development, and accountability legible enough for the institution to govern itself.
Executive activity often appears irrational when interpreted solely as operational production. Attending meetings, maintaining relationships, revisiting decisions, or asking others to prepare work may be wasteful; it may also be the substance of coordination, judgment, and institutional reproduction. Pancakes should preserve the possibility of multiple interpretations rather than assigning a managerial meaning from behavioral traces alone.
The reciprocal warning is equally important: labeling conduct “executive work” must not shield it from criticism. Different institutional horizons explain why behavior may look different from different positions. They do not make authority self-justifying.
The argument can be summarized in the following propositions:
Management does not regenerate automatically when an office changes hands. A title can be transferred in a moment; the capacity to inhabit it is accumulated through exposure, interpretation, practice, correction, trust, and gradually enlarged responsibility.
This makes apprenticeship central to institutional continuity. The process may be formal, as in professional training, or embedded in ordinary organizational relationships. An executive assistant who appears to be handling calendars and correspondence may also be learning how information enters the executive office, how decisions are framed, how authority is delegated, and how the institution understands itself. A later promotion may reveal a developmental function that was present but publicly illegible all along.
Yet reproduction is not inherently virtuous. The same mechanisms that transmit judgment can transmit prejudice, secrecy, loyalty tests, and institutional blindness. The same proximity that enables learning can exclude everyone outside a favored circle. The same informal delegation that prepares a successor can create unaccountable shadow authority.
The institutional problem is therefore not simply to preserve management. It is to reproduce managerial capability while keeping authority legitimate, development accessible, judgment corrigible, and succession accountable.
The companion relationship between the two papers can now be stated precisely:
Management fails when a role is enacted pathologically.
Management also fails when an institution cannot reproduce the capacity to enact that role well—or when it reproduces the pathology along with the role.