When Does a Delegate Become a Ruler?

Delegation is not a surrender of sovereignty but a contract for a function

“How Do We Make Decisions Without a Permanent Ruler?” already established the basic architecture of decision-making without a permanent ruler, and “How Can Communities Cooperate Without a Single Center?” carried it across communities: a clear mandate, a bounded domain, reporting and a requirement to return fundamental changes to the principals. This article does not repeat that design. It studies the moment when a good design begins to wear down.

Delegation makes sense because an agent may have time, expertise, data or operational capacity that a larger group lacks. A specialist may run a treatment plant better, a negotiator may coordinate a contract faster, and an executive board can process hundreds of small decisions that a community cannot vote on every day. But transferring a task is not the same as transferring an unlimited right to decide.

A delegate remains a delegate while it is clear whose function is being performed, under which mandate, within which boundaries, and when the delegate must return for new authority.

Political science and economics often describe this tension as a principal–agent problem. The principal — voters, members, a community or a founding body — entrusts a task to someone else. The agent — a representative, board, administration or expert service — performs it on the principal’s behalf. The difficulty is that their interests need not be perfectly aligned and that the agent often acquires more information through the work.

Information asymmetry is not itself an abuse; it is often the reason delegation exists. Yet the same expertise, document access and daily presence can create an information monopoly. The agent begins to determine what the principal can see, how the alternatives are framed and which problems reach the agenda. Oversight then fails not because nobody votes, but because people vote on a world already filtered by the person they are supposed to oversee.

Accountability therefore cannot rest only on final outcomes. It needs access to reasons, data, alternatives and decision traces. Without them the principal cannot distinguish legitimate expert discretion from silent drift away from the mandate.

A delegate becomes a ruler when the mandate stops constraining discretion

The most useful boundary is not the title of the office but the structure of power. A board chair can be a tightly bounded delegate, while a formally modest coordinator can become a de facto ruler if that person controls information, money, access to procedure and the process of their own reappointment.

The transition becomes especially likely when four elements combine: the mandate is vague enough for the agent to interpret it alone; the domain of decision expands without fresh explicit authorisation; the principal lacks an independent information source; and removal or non-renewal is so difficult, costly or politically risky that it becomes almost theoretical.

A ruler does not need the ability to do everything. It is enough to be able to set important conditions for others on a durable basis without effectively returning to those on whose behalf the authority is exercised whenever discretion expands.

At this point an important correction is needed to the intuitive formula that a representative should simply execute the voters’ will. In European representative democracy, parliamentary mandates are generally free, general and not subject to political revocation during the term. A parliamentarian is not normally legally bound by instructions from one constituency or party leadership, because after election the role includes judging the wider public interest and responding to information that may not have existed at election time.

A binding or imperative mandate solves a different problem: it reduces agent discretion and preserves a tighter connection to the principal. That may be very sensible for an inter-community negotiator, a cooperative delegate or a representative with a specifically enumerated task. It can also become dangerous if the agent is forbidden from exercising any judgment, or if the real commander is not the community but a narrow party leadership.

THY-REALITY therefore does not seek one formula for every level. The more technical, specific and reversible the task, the easier it is to bind the mandate. The broader, less predictable and more rights-sensitive the decision, the more room judgment may require — and the stronger ex post accountability must become.

A party, board or apparatus can become a second principal

In theory, a representative is the people’s agent. In practice there are often several principals at once: voters, organisational members, a party, donors, a coalition, a professional body or the group that controls renomination. Political science therefore also studies multiple-principal or common-agency problems: one agent is answerable to several centres whose demands do not always align.

PACE points to this exact tension. European parliamentarians are supposed to act freely as a matter of law, while party discipline, nomination procedures and the threat of exclusion can generate strong practical pressures. This does not mean parties are inherently undemocratic; parties reduce coordination costs and help voters make sense of political alternatives. The problem begins when a representative’s career becomes more dependent on the internal apparatus than on the people to whom decisions should be explained and justified.

Every audit of delegation should therefore ask: who can actually reward, punish, renominate or block the agent’s next career step? The formal principal and the effective principal are not always the same person or group.

Professionalised authority can build capacity — and distance

Long service in a public or communal function can produce expertise, institutional memory, negotiating skill and understanding of complex procedures. Amateurism is not automatically a safeguard against power; an inexperienced representative can become even more dependent on permanent staff, lobbyists or specialists whom they cannot effectively challenge.

At the same time, a political career creates its own incentive environment. A person whose livelihood and status depend for decades on office, networks, nomination and internal advancement will rationally begin to protect that position as well. This need not be corruption; it can be ordinary human adaptation. But when success inside the system matters more than a verifiable relationship with the principal, the risk of social and informational distance grows.

Recent research on political professionalisation cautions against simple conclusions: different measures produce mixed patterns, and the label ‘professional politician’ does not by itself prove democratic decline. A better test is whether professionalism increases capacity without simultaneously reducing accountability.

Budget, agenda and information can matter more than the formal vote

A delegate can remain formally bounded while still gaining substantial power through three quiet resources: agenda, budget and information. Whoever determines which option reaches a vote already shapes the outcome. Whoever prepares the budget determines which functions are practically possible. Whoever controls reports and expert data defines the frame in which others judge.

Agency drift therefore rarely begins with a dramatic seizure of power. It begins with small precedents: ‘we did this once before’, ‘this is only a technical detail’, ‘there is no need to ask the membership again’. Over time the exception becomes practice, practice becomes competence, and competence begins to generate its own reasons for permanence.

A good system therefore asks more than who casts the final vote. It also asks who prepares the options, gathers the information, controls the resources and can change the scope of the assignment without fresh approval.

Recall can be a safeguard, but it is not a universal cure

An intuitive response to the agency problem is recall: if the agent no longer serves the principals, replace them. In some local, cooperative or contractual roles this can be a powerful safeguard. The Venice Commission recognises that recall can strengthen accountability, while also warning about instability, manipulation and permanent campaigning.

It is therefore important to separate political recall for lost confidence from removal for unlawful conduct. Legal violations belong to appropriate independent procedures; political distrust may, in systems that permit it, activate a political mechanism. If the two are collapsed, recall can become punishment for an unpopular but lawful decision, or a substitute for legal responsibility.

This article therefore does not prescribe recall for every representative. The deeper principle is effective corrigibility: in one institution this may mean recall, in another periodic reauthorisation, non-renewal, dissolution of the body, cancellation of a specific delegation, or mandatory referral of fundamental decisions back to the principal.

The same is true of term limits and rotation. A short term can inhibit entrenched personal networks and reduce the chance that office becomes a lifelong status. Excessive turnover, however, can destroy institutional memory, increase dependence on permanent staff and remove people who have proved competent and trustworthy.

Empirical research on U.S. state legislatures reveals another important tension: when legislators are term-limited and can no longer seek re-election, some electoral incentives for effort can weaken. One large study found that such legislators, on average, sponsored fewer bills, were less productive in committees and missed more floor votes. This does not prove that term limits are bad; it shows that removing a career incentive can also remove part of an accountability mechanism.

Rotation is therefore best used as a targeted safeguard for roles with high capture risk, conflicts of interest or excessive concentration of information — not as a magical solution for every institution.

A good mandate contains boundaries, feedback loops and return points

The strongest protection against a delegate becoming a ruler is not permanent suspicion but good institutional geometry. The agent needs enough room to perform the task, while the principal needs several independent paths for seeing, understanding and, when necessary, correcting the agent’s actions.

1. The mandate should state both the task and what is outside the task; expansion of competence requires fresh authorisation.
2. Important decisions should leave a trace: proposal, reasons, vote, contract, expenditure and result.
3. The agent should not be the only source of information about its own performance; independent data, audit or verification must be possible.
4. Conflicts of interest, gifts, outside employment and future career ties should be disclosed and proportionately constrained.
5. A fundamental change of goal, major new financial burden or new coercive power should automatically return to the principal for a fresh mandate.
6. The mandate needs an expiry date or regular reauthorisation point; automatic renewal should not quietly become permanence.
7. There must be a workable process for non-renewal, replacement or recall where appropriate, without destroying the function itself.
8. Where possible, policy preparation, execution and oversight should be at least partly separated so that the same agent does not evaluate itself.

These are not anti-expert or anti-representative rules. They are the rules that make it possible to trust representatives without requiring blind trust.

Practical audit: does our delegate still serve the mandate?

A community, cooperative, federative body or other organisation can periodically review its relationship with delegates through a set of direct questions:

1. Can we state in one paragraph what the delegate was authorised to do?
2. Which decisions may the delegate take alone and which must return for confirmation?
3. Have the delegate’s powers expanded since appointment — and who explicitly authorised that expansion?
4. Can members independently verify the key information behind the delegate’s decisions?
5. Who controls the agenda, budget, staffing and access to documents?
6. To whom is the delegate practically more dependent: members, party, administration, funder or a professional network?
7. Are conflicts of interest and future career connections visible?
8. When is the mandate reauthorised, and what happens if reauthorisation is refused?
9. Can the person be replaced without the function itself collapsing?
10. Would we accept the same rules if tomorrow the office were held by someone we trusted much less?

The final question is often the strongest. An institution is safe only when its limits do not depend on the assumption that the ‘right person’ currently holds office.

This article has examined the slow failure mode of delegation: more information, more discretion, greater professional dependence and increasingly difficult replacement. The next article will examine a faster version of the same problem — the moment when authority is deliberately expanded because of a crisis.

In a fire, epidemic, flood, attack or other serious emergency, faster and more centralised decisions can be rational. But this creates a new dilemma: what ensures that temporary authority truly ends when the reason for the exception is gone?

“Emergency Powers: Why Temporary Power Tends to Stay” will therefore examine emergency powers, sunset clauses, independent review and the mechanisms that return a system from exception to ordinary rules. This article supplies the prerequisite: authority needs an owner, a boundary and a route back before the moment arrives when limiting it will be hardest.