When Does Reform Actually Happen?

How a proposal acquires a sponsor, enough support, a lawful decision, and an institution able to carry it out

Elias Kunnas

Corpus frame

The corpus applies one lens to many domains: what mechanisms produce the outcome? It shares four methodological commitments and one explicit directional commitment. Each linked page argues for its part; the links are derivations and disputes, not evidence inherited by every page. The directional commitment does not by itself settle system boundary, distribution, sacrifice, or institutional authority.

Mechanisms are what act. Incentive gradients, selection pressures, feedback loops, and capital stocks produce the distribution of outcomes. Intentions, labels, official categories, and stated values are evidence about mechanisms, or are themselves coordination mechanisms. They are not causal substitutes. — Mechanism Realism · Only Selection
The reference telos is sustained flourishing. The broadest achievable adaptive safety margin over deep time — not the continuity of any incumbent state, coalition, institution, or doctrine. A mechanism's own stated goal can still serve as a local proof obligation — showing that its incentives defeat even the purpose it claims is a bounded finding — but meeting that goal establishes nothing about the margin. — Flourishing Is Maximum Safety Margin
Law, rights, legitimacy, democracy, markets, and sovereignty are mechanisms under evaluation. They are constraints, carriers, or proxies inside the analysis. None is a terminal value or a boundary of what is real. Treating one as terminal ends the mechanism search before it starts. Evaluation carries current function, replacement cost, path dependence, uncertainty, capture risk, reversibility, and who bears model error into the ledger. — The Stack · Mechanism Space
Optimization is a system function. A civilization has to build, exercise, and revise metamechanisms that search mechanism-space, discard dominated options, install, observe effects, and repair under uncertainty. Not running that loop leaves margin unrealized, and that is itself the failure. No single component — analyst, model, or institution — is presumed to contain a global optimum; the capacity is a property of the system. — From Telos to Policy · The Three-Layer Architecture
Uncertainty is preserved, not spent. Partial orders, binding constraints, unknowns, and residuals stay explicit. An unmeasured effect is not a favorable default. — The Compression Paradox · Cargo Cult Epistemology

Each essay bears its own evidence. Links carry definitions, derivations, applications, and disputes; they do not transfer proof. Criticism is answered on its substance.

Where each commitment is derived

A legislature can have the legal power to create an institution without there being a collective actor willing and able to make that decision. Reform becomes possible when a bounded proposal has someone prepared to advance it, the necessary actors can support or tolerate the same operative text, a lawful means of deciding it is available, and a named institution has the authority and resources to carry it out. Publicity, crises, expert reports, elections, referendums, bargains, and demonstrations can change those conditions. None supplies the whole arrangement by itself.

Standard objections addressed in this essay

“A parliamentary majority can simply vote for it.” — §I (the vote is the final event, not an explanation of how it became available)
“Public pressure can force reform.” — §V (pressure matters through identifiable actors, costs, and decisions)
“Crises really do cause reform.” — §VI (a crisis changes conditions but does not choose or implement a proposal)
“This is just policy entrepreneurship.” — §II (the ingredients are established; the contribution is a proposal-level audit)
“Institutional change can be gradual and have no single vote.” — §II (the scope is deliberate reform requiring an identifiable public authorization)

I. Two Hundred People Who Could

Finland has 200 members of Parliament. In a legal sense, a majority can create a new public institution. It does not follow that there is a political actor called “the 200 members” with one objective, one agenda, or one willingness to bear the cost of acting.

“They could fix it if they wanted” compresses the entire reform problem into a hypothetical shared preference. It treats formal capacity as organized capacity. The same move appears whenever a problem is assigned to “the government,” “the public,” “the parties,” or “decision-makers.” Each label names a set of people who could contribute to a decision. None identifies who will advance this proposal, who must accept it, which procedure can authorize it, or which institution will carry it out.

A majority becomes relevant only around a particular text. Before the vote, someone must define what is being created, obtain a place on an agenda, negotiate the terms, assemble enough support or non-opposition, and ensure that an accepted proposal can be implemented. These tasks may be performed by one unusually powerful actor, by several institutions, or by a temporary alliance. Until they are performed, the majority exists only as arithmetic.

Public knowledge can coexist with institutional immobility. Agreement on a diagnosis does not select a remedy. Agreement on a remedy in the abstract does not produce agreement on its powers, location, budget, appointments, or effect on existing discretion. The useful question is narrower than “who has the power to fix this?”

What would have to become true for this particular reform to reach an authoritative decision and survive into operation?

II. Reform Stories Mix Different Kinds of Things

Accounts of successful reform often identify one decisive cause: a crisis, a policy entrepreneur, a grand bargain, a referendum, external pressure, a technocratic government, a court judgment, a pilot, or a prepared proposal waiting for a window. These labels describe different parts of political change, not rival answers to one question.

A crisis can change perceived stakes, time horizons, blame, and the cost of delay.
A policy entrepreneur can supply attention, continuity, negotiation, and personal risk-bearing.
A bargain can align actors whose assent is required.
A referendum supplies a decision rule and a source of legitimacy where such a procedure exists.
An external constraint can change domestic payoffs and let politicians attribute an unpopular decision to an outside obligation.
A technocratic government temporarily relocates initiative and executive authority.
A court judgment can compel action under an existing legal commitment.
A pilot can reduce uncertainty and create an experienced group able to continue the work.
A public report can define the problem, provide evidence, or give an advocate cover.

A referendum without an initiative, a text, a campaign, and later implementation does nothing. A crisis without a prepared and politically selectable response can produce delay, symbolic action, or the wrong reform. A champion without agenda access remains an advocate. A pilot without a later decision can become an admired exception that never changes the general rule.

Political science already contains substantial accounts of these ingredients: policy entrepreneurship, veto players and agenda control, institutional entrepreneurship, and the way uncertainty about winners and losers can defeat reforms that might later become popular.¹²³⁴ The claim here is not that these traditions missed political change. Public discussion routinely substitutes one ingredient for a complete explanation.

Within this corpus, Selling Is Futile asks whether a decision structure exists that owns a problem and can acquire an offered solution. Civilization’s Blind Spot shows how reformers have repeatedly performed the work of a missing institution until a durable organization existed. This page specializes the public-reform case in which authority is distributed and no unitary buyer exists: several roles must be assembled around the same decision.

The domain is deliberate institutional reform in a constitutional democracy: a change that requires an identifiable act of public authorization, such as a law, decree, budget decision, parliamentary rule, appointment, or binding administrative mandate. Drift, reinterpretation, layering, diffusion, and changing norms without one defining authorization require a different account.

III. Begin with the First Decision

A reform becomes politically concrete when its first state-changing decision can be stated in one sentence. “Improve accountability” is an objective. “Create an independent office” is still incomplete. A decision identifies at least the institution, its initial task, its legal basis, its responsible home, its resources, and the date or event by which it must act.

The first decision need not install the final architecture. Bounded does not mean small: a body may decide a package, but the package must be definite enough to accept, reject, amend, authorize, and fund. It may establish a time-limited office, authorize a trial, impose a response duty on a defined class of decisions, assign preparation to a ministry, create a budget line, or require Parliament to revisit the question after specified evidence has been produced. The important property is that the decision changes institutional reality and leaves a defined next decision rather than another general aspiration.

Broad agreement often disappears at the point where a proposal starts reallocating discretion. Many actors can support “better impact assessment.” They may disagree over whether an evaluator is independent, whether its opinion delays legislation, whether ministers must answer it, whether Parliament can override it by ordinary majority, which proposals fall within its scope, who appoints its members, and which budget pays for it. Before an operative text exists, apparent support is cheap. After it exists, the required concessions, losses, and safeguards become visible.

A bounded first decision also changes the search. Instead of asking for general support for a worldview, the reformer can ask who has authority over this decision, what each pivotal actor would gain or risk, and which formal procedure could carry it.

IV. The Conditions of an Adoptable Reform

A live reform attempt can be examined through five questions. They are dependencies, not a fixed chronological sequence. The proposal may change to attract a sponsor; the available legal form may determine which actors must agree; a pilot may create both evidence and a future constituency. One person or institution may fill several roles.

1. What exactly is the first decision? The proposal must be bounded enough that an authorized body could accept, reject, or amend it. A diagnosis, aspiration, framework, or institutional sketch may be valuable, but it is not yet the decision. Evidence includes a draft provision, decree, budget decision, parliamentary motion, appointment proposal, procurement, or formally described trial with a defined scope and endpoint.

2. Who will advance it, and through which agenda? Someone must spend scarce time, reputation, staff effort, and bargaining capital on the proposal. The beneficiary is often not this person. A real sponsor is identifiable by a next action: requesting preparation, securing a hearing, negotiating inclusion in a programme, drafting the instrument, assigning staff, or placing the matter before a body that can decide it. Public approval without such an actor is an audience, not sponsorship.

3. Whose support or non-opposition is required? Formal voting rules provide only part of the answer. Agenda controllers, coalition partners, committees, ministries, constitutional reviewers, budget authorities, organized interests, and implementing bodies may each be able to stop, dilute, delay, or substitute the proposal. The relevant coalition is proposal-specific. Its members need not share a theory of government. They must be able to accept the same operative text. A slogan that conceals incompatible expectations can assemble initial support and then fail during drafting or implementation.

4. What formal means can authorize it? The available procedure determines who participates, what majority is needed, how quickly the matter can move, and how durable the result will be. A government decree, ordinary statute, constitutional amendment, budget decision, parliamentary standing order, referendum, court order, or treaty obligation creates a different political problem. Choosing the strongest legal form is not always optimal. A reform may be achievable by decree and impossible by statute, or achievable as a trial and impossible as a permanent office. The choice has to be made rather than hidden inside the word “reform.”

5. Who can carry it out after acceptance? Authorization without administrative capacity produces a nominal reform. The responsible institution needs authority, staff, money, information access, deadlines, and a way to resolve conflicts with existing bodies. It also needs a defined event at which continuation, revision, expansion, or termination will be decided. Actors may agree to a law precisely because they expect weak implementation. An existing institution may have relevant competence but no mandate or resources to assume the new task.

These five questions do not guarantee success. They make failure more local and testable. An attempt with no sponsor is in a different state from one with a sponsor but no acceptable legal form. “There is no political will” erases that difference.

V. The Threshold Must Belong to Someone

A common explanation says that reform occurs when the cost of inaction exceeds the cost of change. After a reform succeeds, one can say the threshold was crossed; after it fails, one can say it was not. The statement acquires content when the costs are attached in advance to actors whose choices matter.

For any proposed threshold, ask: whose expected cost changes; through what event; which decision the cost affects; on what timescale; and compared with which alternatives. The relevant actor may prefer symbolic action, delay, substitution, blame-shifting, or a narrower reform over both full adoption and open refusal.

A rising debt-to-GDP ratio is not itself a political threshold. It becomes one if it triggers borrowing costs, treaty consequences, a coalition commitment, voter attribution, or another event that changes a pivotal actor’s options. A demographic forecast may describe enormous social loss while imposing almost no near-term cost on anyone capable of acting.

This is also the precise role publicity can play. Visibility may increase electoral or reputational cost, coordinate previously isolated supporters, narrow plausible denial, or give a sponsor evidence with which to bargain. Visibility that changes no pivotal actor’s expected choices remains information.

VI. What Familiar Reform Forces Actually Supply

The question “which reform strategy works?” encourages lists of supposedly alternative methods. A better question is which missing condition a method can supply in this case.

Familiar explanation	What it can change	What it still leaves to be supplied
Publicity or common knowledge	Attention, attribution, coordination, reputational cost	A sponsor, proposal-specific support, authorization, implementation
Acute crisis	Stakes, time pressure, blame, tolerance for exceptional action	Selection of a remedy, an operative text, competent execution
Expert report or independent analysis	Diagnosis, evidence, comparison, a draft proposal	Political risk-bearing and a place on an authoritative agenda
Policy entrepreneur	Continuity, framing, negotiation, personal investment	Required assent, lawful authority, administrative capacity
Grand bargain	Support across pivotal organized actors	A bounded decision, excluded interests, implementation and later correction
Referendum	A decision rule, public mandate, possible bypass of legislative bargaining	Initiative access, campaign, text, later administration
External condition	Changed payoffs, deadlines, political cover	Domestic ownership, lawful transposition, implementation
Court judgment	A binding decision under existing law	Scope beyond the case, resources, sustained administration
Trial or parallel institution	Feasibility evidence, trained people, practices, a constituency	Public authority and a decision to continue or expand
Technocratic interlude	Concentrated initiative and temporary executive capacity	Durable legitimacy, electoral support, survival after ordinary politics returns

A reform window is the temporary condition in which enough of these contributions coincide around one proposal. The window does not act. People and institutions act differently because the window changes their options, incentives, information, or authority. A prepared proposal can reduce the work required during a short opening. It still needs selection, support, authorization, and execution.

VII. A Finnish Case: The Legislative Evaluation Council

Finland’s Legislative Evaluation Council can be reconstructed as a sequence of public acts rather than attributed to a national crisis.

In December 2014, the government steering-system project known as OHRA proposed an expert unit for impact assessment. A preparation group followed with a proposal for a temporary evaluation board in April 2015. Parliament had required the government to take measures to establish a unit dealing with legislative impact assessments. The programme of Prime Minister Juha Sipilä’s government then committed to creating an impact-assessment body in connection with the government. The Government issued Decree 1735/2015; it entered into force on 1 February 2016. The first council began its term on 15 April 2016.⁵

No single item explains the result. The OHRA proposal did not appoint a council. Parliament’s statement did not write the decree. The government programme did not provide a functioning secretariat. Together they produced an authorized institution.

The later history is a different decision. The programme of Prime Minister Petteri Orpo’s government called for strengthening the council by regulating it in statute. Government Proposal HE 137/2025 would have added the council to the Act on the Government, preserving its mainly advisory role while placing its central provisions at the level of law.⁵

The Constitutional Law Committee proposed deleting that section. It accepted the importance of impact assessment, but considered regulation by decree constitutionally sufficient. It raised concerns about increasing the legal regulation of legislative preparation, preserving the government’s political role in preparing bills, and placing an independent advisory body in that particular statute.⁶ Parliament enacted Act 150/2026 without the proposed provision. The council continues under the decree-level arrangement.⁷

This is not a story of reform succeeding once and then simply “losing political will.” The 2015–2016 decision was whether to establish the body through a government decree and place it at the Prime Minister’s Office. That decision succeeded. The 2025–2026 decision was whether to entrench the body in statute within a particular legal structure. That decision encountered a specific institutional objection at a body with the power to reshape the bill, and the proposed provision was removed.

Creation, authority, resources, and legal durability are separate political decisions. Broad agreement on a goal does not imply agreement on institutional form. A decree-level creation and a statutory entrenchment present different questions to different actors. Reconstructing the adoption history replaces “the system eventually acted” with who proposed, required, committed, authorized, housed, and later limited what.

VIII. An Audit for a Live Reform Attempt

Before asking whether a reform is popular, inevitable, or historically validated, write down answers to these questions:

What exact first decision would change institutional reality?
Who can place that decision before a body authorized to make it?
Who is prepared to spend time, reputation, staff effort, and bargaining capital to advance it?
Whose assent or non-opposition is required, formally and in practice?
Why can each pivotal actor accept the same operative text?
What law, decree, budget act, parliamentary rule, court process, or other formal means can authorize it?
Which institution will receive the authority, staff, money, information, and deadlines needed to carry it out?
What later event will decide whether the reform continues, changes, expands, or ends?

A usable answer names actors and acts. “The public demands it” does not say who places it on an agenda. “Parliament can decide” does not identify the proposal or the route to a vote. “A crisis will force action” does not identify the action. “A pilot proves it works” does not confer authority. “The government will implement it” does not name the responsible office, budget, or deadline.

The audit is a status test, not a claim that absent pieces can never appear. The present attempt cannot culminate until the missing condition changes. If the first decision is undefined, write it. If the sponsor is absent, more technical detail may have low marginal value. If one pivotal actor expects a concentrated loss, broader awareness may strengthen opposition as easily as support. If the legal form is unavailable, the proposal may need a different first step. If implementation has no home, winning the vote may only create a paper obligation.

IX. The Mechanism Authority Under Its Own Test

The Mechanism Authority is a proposal for an independent public function that tests the causal design of major policies before adoption, follows outcomes afterward, and requires decision-makers to answer identified failures. Its design work addresses what the institution would do after it exists. That is different from an account of how it comes to exist.

Under the audit, the current proposal remains incomplete. Several possible legal instruments have been identified, but the first state-changing decision has not been fixed. No political office, party, committee, or ministry has committed to advancing that decision. The required support or non-opposition has not been mapped around one operative text. Possible administrative homes exist, but none has accepted responsibility. Analyses and prototypes can demonstrate methods; they do not grant public authority. The honest status is therefore: the institutional design is substantially specified; the adoption arrangement is not.

One possible first decision is narrower than the complete institution: a fixed-term mechanism-assessment trial in a named administrative home, for a bounded class of major reforms, with public assessments before and after enactment, a written ministerial response, a budget and staff, and a predetermined decision on continuation or termination. This is only a candidate. It can be tested against Finnish authority: who could request it, by what legal and budgetary route, who would see it as support or as an intrusion, and what powers would make the trial worth running without making the initial coalition impossible.

The Legislative Evaluation Council shows the level of specificity an adoption account requires. It does not show that the Mechanism Authority would follow the same path. External work still matters: a working method can reduce uncertainty; published analyses can reveal demand, train people, and give a future sponsor evidence. Each is preparatory unless it changes one of the conditions above. The proposal should therefore be developed on two separate tracks: the design of the institution and the construction of its first authoritative decision. Progress on one cannot be counted as progress on the other without an explicit connection.

X. Reform Is Organized Authority Around One Decision

Public debate usually begins at too broad a level. The country is failing. The system is corrupt. Everyone knows. Parliament has the power. A crisis is coming. A better institution has been designed. All of these statements can be true while no reform is in motion.

The unit that moves is smaller: one bounded decision, advanced by identifiable people, accepted or tolerated by the actors who can stop it, authorized through an available procedure, and carried out by a responsible institution. The arrangement can be democratic, judicial, administrative, external, or mixed. It can be assembled slowly or compressed by crisis. Its participants can support the same decision for different reasons. This does not make reform easy. It makes the missing work visible.

“The cost of inaction eventually becomes too high” becomes: whose cost changes, through what event, and which decision does that make them take? “The public must wake up” becomes: which pivotal actor will behave differently when it does? “The government could fix it” becomes: who will put which text before which body, with whose support, and who will implement the result? A reform window exists when enough of those answers become favorable at the same time. The window is not the cause that acts. It is the temporary opportunity for a prepared and organized decision to be made.

This essay is subject to its own argument. Publishing an audit can clarify why reform attempts stall and can improve a proposal. It does not create a sponsor, a coalition, a formal decision, or an implementing institution. Its consequence begins when someone uses it to name the next missing act in a real attempt and then performs that act.

Related:

Civilization’s Blind Spot — why successful correction has repeatedly depended on people temporarily performing the work of a missing institution
Selling Is Futile — why an offer fails when no decision structure owns the problem and can authorize acquisition
The Freedom of Inconsequence — why expression and association remain harmless when they are disconnected from operative consequence
Bad Equilibria Are Not One Thing — the broader diagnostic that distinguishes a missing joint move from other failure classes
The Procedural Object — how a claim becomes admissible to an existing forum
The Fourth Branch — the wider institutional design that the Mechanism Authority instantiates

Sources and Notes

1. Policy entrepreneurship. Michael Mintrom and Phillipa Norman, “Policy Entrepreneurship and Policy Change,” Policy Studies Journal 37(4), 2009, pp. 649–667. Actors who invest resources in change and connect problems, proposals, and political opportunities. Consulted: journal record and abstract.

2. Veto players. George Tsebelis, “Veto Players and Institutional Analysis,” Governance 13(4), 2000, pp. 441–474. Whose agreement is required, and how agenda control shapes available outcomes. Consulted: journal record.

3. Institutional entrepreneurship. Julie Battilana, Bernard Leca, and Eva Boxenbaum, “How Actors Change Institutions: Towards a Theory of Institutional Entrepreneurship,” Academy of Management Annals 3(1), 2009, pp. 65–107. Actors from the emergence of a change project into implementation. Consulted: journal record.

4. Uncertainty and status-quo bias. Raquel Fernández and Dani Rodrik, “Resistance to Reform: Status Quo Bias in the Presence of Individual-Specific Uncertainty,” American Economic Review 81(5), 1991, pp. 1146–1155; NBER working paper 3269. Uncertainty about who will gain and lose can defeat reforms that might become popular after adoption. Consulted: working-paper version.

5. Founding sequence and 2025 statutory proposal. Government Proposal HE 137/2025, especially §1.1. OHRA December 2014; April 2015 preparation group; Parliament’s statement EK 46/2014 vp; Sipilä government programme; Decree 1735/2015; 1 February 2016 entry into force; first council 15 April 2016; Orpo programme commitment to statutory regulation. Consulted: Finlex HTML, 12 September 2026.

6. Deletion of the proposed statutory section. Constitutional Law Committee Report PeVM 6/2025 on HE 137/2025, especially paragraphs 7–14 and the detailed proposal to delete 26 a §. Decree-level regulation constitutionally sufficient; concern about juridifying legislative preparation; protection of the government’s political role in preparing bills; the provision does not belong in the Act on the Government. Consulted: Edilex HTML, 12 September 2026.

7. Enacted law without the council provision. Act 150/2026 amending the Act on the Government. The enacted law does not contain a section on the Legislative Evaluation Council. Consulted: Finlex HTML, 12 September 2026.

8. Continuing operation. Prime Minister’s Office, tasks, methods, and composition of the Legislative Evaluation Council. The council continues under the decree-level arrangement. Consulted: office page, 12 September 2026.

Scope. The five-question audit is an integrative diagnostic for deliberate, proposal-specific public reform. It is not a replacement for policy-process research, a universal causal model of institutional change, or an account of gradual change without an identifiable authorization event. The Legislative Evaluation Council is a reconstructed pair of decisions, not a prevalence claim about Finnish reform.

Source check: 12 September 2026.