CORE PATH Stop 47 / 106

Natural Law October 16, 2021 Updated: September 11, 2026 14 min read THY-REALITY

Institutional Capture: When the System Starts Serving Itself

How does an institution created for a shared function gradually begin defending its own interests? This article separates capture from corruption and examines regulatory capture, local elites, nepotism, revolving doors, procurement, information control and organisational self-preservation.

“Emergency Powers: Why Temporary Power Tends to Stay” showed why emergency power needs an ending built into its beginning. But institutions can develop interests of their own even without a state of emergency. An agency, board, party, company, cooperative, municipality, regulator or other organisation is created to perform a function; over time, however, it may begin defending its budget, status, personnel and authority so persistently that the original purpose shifts from the reason for its existence into a justification for preserving the institution itself.

Institutional capture occurs when a system's decisions are no longer directed mainly toward the people or function for which the system was created, but are repeatedly redirected toward the benefit of a narrower group, an internal network or the organisation itself. Corruption can cause this, but it is not required. Capture may also grow through legal lobbying without counterweights, staffing networks, dependence on one expert perspective, privileged information access, procurement, revolving doors or simple organisational self-preservation.

This article is therefore not an argument that every institution is corrupt. It is an argument about feedback loops. If a system cannot detect when its internal logic has drifted away from its purpose, it can behave rationally for its insiders while performing badly for the community. The task is not to abolish organisation, but to preserve the distinction between an institution as a tool and an institution as a new master.

When an institution replaces its purpose with its own survival

Every durable organisation needs some degree of self-preservation. It must employ people, retain knowledge, plan budgets, protect infrastructure and survive leadership changes. Without that capacity it could not perform its function. The problem therefore does not begin merely because an institution pays attention to its own continuity.

The problem begins when the relationship reverses. Instead of preserving the organisation so that it can perform the function, the function is interpreted so as to justify preserving the organisation. A programme that no longer delivers results is not ended but re-described. An authority that could be devolved or opened to competition is declared too sensitive. Criticism begins to be treated as an attack on the mission itself.

This is this article's first warning sign: an institution is healthy only while it can still accept that one of its tasks may be finished, unnecessary or better performed elsewhere. A system that can only expand can no longer credibly demonstrate that it serves its function rather than itself.

Corruption commonly involves abuse of entrusted power for private gain. Institutional capture is a broader pattern. The OECD describes policy capture as the repeated direction of public decisions away from the public interest toward a specific interest. This can happen through illegal means such as bribery, but also through legal channels when access, information and influence become severely unbalanced.

Nor is every bad decision evidence of capture. An institution can make mistakes, lack knowledge or operate through a badly designed process. What matters more for capture is the pattern: the same network repeatedly receives access, contracts, exceptions, appointments or favourable interpretations of rules, while opposing interests struggle to reach the process, obtain information or use an effective remedy.

This distinction avoids two opposite errors. The first is cynicism, in which every contact between an interest and a decision-maker is treated as corruption. The second is naivety, in which everything is considered healthy unless someone can prove a cash bribe. Capture can be legal, gradual and institutionally normal — which is precisely why it can be hard to see.

Regulatory capture: when the overseer starts seeing through the eyes of the regulated

The best-known form is regulatory capture. A regulator exists to protect a broader function — competition, safety, the environment, service quality or fair access. But if information, personnel and expertise come almost exclusively from the industry being regulated, the regulator's perspective can gradually narrow into the perspective of regulated actors.

This does not require bad faith. Regulated firms often genuinely know the technical details best. The problem arises when they also become nearly the only source of data, candidates for expert posts, draft rules and explanations of what is feasible. At that point the boundary between expert consultation and dependency can disappear.

The safeguard is therefore not to exclude industry from discussion, but to create plurality of information and access: competitors, users, workers, independent experts, local communities and other affected groups need a real opportunity to contribute evidence and challenge assumptions. A regulator that hears only one side can become its extension while remaining formally independent.

THY-REALITY often argues for subsidiarity and for placing decisions closer to those who live with their consequences. This article must state the other side just as clearly: decentralisation by itself does not remove capture. The World Bank explicitly lists elite capture, patronage, corruption and opaque local decision-making among the risks of decentralisation.

A small community may have shorter information chains and stronger personal accountability, but it can also contain long-standing family, business or political networks that are harder to challenge precisely because everyone knows everyone. A local monopoly over land, contracts, permits, media or employment can create intense dependency without a large bureaucracy.

Subsidiarity protects against distant centralisation only when it is accompanied by open access, appeal, cross-checks and exit. If the local level becomes a closed caste, proximity no longer means accountability; it simply means that capture is closer to home.

Nepotism and patronage: when loyalty displaces criteria

Staffing is one of the quietest routes to institutional capture. If positions gradually go mainly to relatives, political allies, friends or members of the same inner circle, the organisation changes before its formal rules do. The OECD emphasises merit-based recruitment because transparent and objective criteria reduce room for patronage and nepotism.

The problem is not only that a less capable person may be selected. More important is the shift in loyalty. If careers depend on personal allegiance, it becomes dangerous to challenge the person who controls advancement. Internal reporting, professional dissent and the institution's ability to correct itself then weaken.

A healthy organisation does not need a mythology of perfect neutrality, but it does need criteria known in advance, open procedures, traceable reasons and a route for challenge. Where the answer to 'why did this person get the post?' is only 'because we trust them', the space for capture is already much larger.

Movement between public, private and civil-society roles is not inherently problematic. It can transfer valuable knowledge, improve understanding of technology and bring practical experience. The OECD explicitly recognises these benefits. The same movement, however, can create conflicts of interest when an official regulates an organisation today and seeks employment from it tomorrow, or when a former regulator immediately begins lobbying former colleagues.

The right questions concern timing, subject matter and traceability. Did the person participate in decisions from which the new employer will directly benefit? Do they hold confidential information? Is there an appropriate cooling-off period, recusal from specific matters or public declaration of relevant interests?

A revolving door is a risk, not proof of guilt. Good rules do not punish mobility; they prevent public office from becoming an investment in future private gain or a private position from buying privileged entry into public decision-making.

Contracts, budgets and procurement as the soft infrastructure of capture

A large part of institutional power does not appear in statutes but in contracts. Who receives maintenance, consulting, software, construction, supply, concessions or long-term service agreements? The OECD and UNODC identify public procurement as especially exposed to conflicts of interest, favouritism, collusion and corruption because of large financial flows and close interaction between public and private actors.

Capture can begin before a tender opens: the need is described to fit one supplier; technical specifications become artificial barriers; exceptions to competition become routine; later, the contract is repeatedly extended because switching suppliers has become too expensive. The organisation gradually becomes dependent on the same circle of contractors.

Safeguards include criteria set in advance, open competition, interest declarations, traceable contract amendments, independent review and a right of challenge. Where the money flow cannot be examined, institutional loyalty is also difficult to examine.

Who controls the information can also control the problem

Institutions decide through information, so information architecture is one of the main routes to capture. Who decides what data are measured? Who writes the expert assessment? Who determines which risk matters and which alternative is 'unrealistic'? If the same inner circle controls the problem definition, the data, the success metric and the evaluation of its own work, formal oversight may exist while substantive oversight remains weak.

Information capture does not require falsification. More often it works through selection. Poor results are interpreted as evidence that more resources are needed, while good results prove the programme's success. Metrics measure activity instead of outcomes. Independent evidence becomes 'not comparable', while critics are said not to understand the complexity.

That is why open data, external audit, methodological transparency and the right of affected people to challenge the official picture matter. “How Do We Measure Whether Decentralisation Is Real?” will go further and ask how concentration of power can actually be measured; this article establishes the prior rule: oversight is not independent if the overseen also defines the only permitted measure of success.

Organisational self-preservation: when the means becomes the end

The most subtle form of capture has no outside captor. The organisation begins serving itself. Its members may sincerely believe in the mission while their salaries, status, professional identity and careers are also tied to the continued existence of the problem they are tasked with addressing. That creates a natural bias against decisions that would reduce the organisation's scope.

This kind of self-preservation is not necessarily a moral failure by individuals. It is a structural conflict. An organisation normally knows its own needs better than the costs it imposes on others. Without an external feedback loop, it can therefore justify a new post, programme or authority more easily than it can justify shrinking itself.

A healthy system needs procedures through which a function can be reduced, transferred, merged, opened to competition or ended without leaving the decision solely to those who directly benefit from its continuation. Self-preservation is human; institutional wisdom means not mistaking it for the public interest.

Safeguards: how a system remains directed outward

There is no single anti-corruption or anti-capture mechanism. In its work on preventing policy capture, the OECD stresses a combination of diverse stakeholder participation, transparency of information, accountability and organisational integrity policies. The combination matters: when one safeguard fails, another route for correction must remain.

In practice this means merit-based staffing, disclosure and management of conflicts of interest, proportionate cooling-off rules, transparent lobbying, open procurement, publication of contracts and amendments, independent audits, whistleblower protection, appeal mechanisms, rotation in particularly sensitive posts, multiple sources of expertise and separation between those who run a programme and those who evaluate its success.

The most important safeguard is corrigibility. An institution must be capable of losing a contract, leader, authority, budget or even its reason for existence when results and needs no longer justify them. A system that cannot lose anything it has once acquired will eventually start defending itself.

Practical audit: does the system still serve its purpose?

For an organisation, regulator, municipality, fund, association, company or other shared system, twelve questions can expose capture risk:

Is the institution's original purpose still clearly defined and verifiable?
Who can independently assess whether the institution is actually achieving that purpose?
Do different affected interests have real access to the process, or does one inner circle dominate?
Are appointments and promotions tied to criteria known in advance, or mainly to personal and political loyalty?
Are conflicts of interest, secondary employment and relevant private interests disclosed and managed?
Do revolving-door transitions include reasonable restrictions concerning specific matters and confidential information?
Are contracts, tenders, exceptions and contract amendments transparent and challengeable?
Can independent actors verify the data and metrics through which the institution evaluates itself?
Does the same network of people, firms or organisations repeatedly appear in appointments, contracts and advisory roles without a clear explanation?
Is there a safe channel for insiders to report problems without destroying their careers?
Can authority, programmes or budgets actually shrink when need declines or results are not delivered?
Who has the practical power to correct, reverse or transfer decisions when capture is detected?

A single negative answer does not prove capture. A repeated cluster of negative answers, however, indicates that the institution's formal purpose is becoming less connected to its real incentives and power flows.

the articles from “Monopoly, Plutocracy and the Concentration of Economic Power” through “Institutional Capture: When the System Starts Serving Itself” have described four different pathways of reconcentration: economic power, delegate drift, emergency centralisation and institutional capture. In all four cases a system can remain 'decentralised' on paper while decisive power is already accumulating in a small number of hands.

The next step therefore cannot be another declaration in favour of decentralisation. We need measures. How many people actually decide? Who controls assets, revenue, data and supply routes? What does exit cost? Is there redundancy, or merely several names for the same dependency?

“How Do We Measure Whether Decentralisation Is Real?” will close this institutional arc with measurable criteria for decentralisation. If power cannot be observed and compared, it can be concentrated for a very long time under the language of freedom, locality or cooperation without anyone noticing that the structure has become exactly what it was designed to avoid.