Part I — Situation overview

At the beginning of July 2026 Transparency International Hungary submitted a public-interest data request to the Hungarian National Bank (MNB): it asked for the data of the central bank’s contracts for consideration and free-of-charge contracts concluded after 1 January 2025. The request — signed by Miklós Ligeti, the organisation’s legal director — concerned the data which, under the general publication list of Act CXII of 2011 on informational self-determination and freedom of information (colloquially: the Freedom of Information Act), have to be published: the type, subject, the names of the contracting parties, the value and the duration of contracts for the procurement of goods, for construction investment, for the ordering of services and for the use of assets, connected with the use of public finance resources and reaching or exceeding a value of five million forints. The central bank does publish today which organisations it grants support of at least five million forints — in the first quarter of this year, for example, the Will Foundation for Women Leaders, the Corvina Foundation and the MNB’s own Foundation for the Common Good, established in 1993, received such amounts — but it does not make its other contracts public (444.hu, 21 August 2026).

The course of the procedure is itself telling. On 24 July the central bank made use of the possibility allowed by law and extended the fifteen-day deadline available for answering by a further fifteen days. Then, on 13 August — a good week after the extended deadline had expired — instead of substantive compliance it gave a refusal: it stated that the publication obligation under the Freedom of Information Act “does not apply to the Hungarian National Bank […], in view of the fact that the MNB is not a budgetary body”, and that it therefore does not even produce a database of that content. On 21 August Transparency International turned to the National Authority for Data Protection and Freedom of Information (NAIH) for legal remedy, on two grounds: on the one hand that after the extension of the deadline there was no longer any lawful way to refuse the request, and on the other that under the Freedom of Information Act the publication obligation applies irrespective of the legal status of the party to the contract, so it is not relevant that the MNB is not a budgetary body (Telex, 444.hu, HVG, Portfolio, ATV, 21–22 August 2026).

The dispute is not new, and that is precisely the point. The central bank’s argument goes back to the 2016 legislative turn according to which assets transferred by the MNB to its own foundations and companies “lose their character as public assets”. János Áder, President of the Republic at the time, sent that amendment to the Constitutional Court before promulgation, and in its Decision 8/2016. (IV. 6.) the panel declared it contrary to the Fundamental Law, because it did not find the necessity of restricting freedom of information constitutionally justified. The decision also recorded the more general proposition that the MNB — as an organisation in exclusive state ownership which directs and supervises the country’s financial system — performs a public task and manages exclusively public money, and therefore owes an account to the public in the interest of transparency and of the cleanliness of public life. That is to say, the argument the central bank uses today already failed constitutional scrutiny ten years ago — it merely returns now in a different legal garb, invoking the absence of the status of a budgetary body.

MIAK’s reading of this rests on a single conceptual distinction, and that distinction holds together the whole line of thought of this post: the independence of an institution is not the same as exemption from accountability. Central bank independence in the EU and domestic legal sense concerns monetary policy decisions — that the central bank may decide on the interest rate, on the management of foreign exchange reserves and on financial stability instruments without political instruction. This protection has nothing to do with whether the public may learn the amount and the partner of a contract concluded for an IT service, for an event or for a consultancy assignment. The blurring of the two questions is not a legal necessity but a choice — and the present case shows precisely that as long as the text of the law does not state the difference, the choice remains a matter of institutional discretion.

Part II — Foundations in the literature

The conceptual framework of the situation comes together in three authors. In his work Controlling Corruption, Robert Klitgaard (American economist, one of the founders of the institutional strand of corruption research) formulates the formula according to which corruption flourishes where monopoly position and discretionary power meet weak accountability — in the present case all three elements are present: the central bank is the sole owner of its own procurement decisions, the criteria of the decision are not public, and the route of external control is a long administrative and judicial procedure. In her volume Corruption and Government, Susan Rose-Ackerman (American jurist and economist, the leading researcher of the institutional causes of corruption) expressly discusses the dilemma of independent regulatory and supervisory institutions: professional autonomy requires protection from political influence, but this protection has to be accompanied by a transparent appointment procedure, conflict-of-interest limits and results-based accountability, otherwise autonomy turns into unaccountability. And the work Governance Matters by Kaufmann, Kraay and Zoido-Lobatón shows that voice and accountability is an independent, measurable governance dimension which is also connected with development outcomes — that is, data publicity is not a matter of mood but an institutional characteristic describable by an indicator. The detailed treatment of the literature — author by author, with quotations — can be found in section 6.4 Literature in detail.

📖 Source: Robert Klitgaard: Controlling Corruption; Susan Rose-Ackerman: Corruption and Government — Causes, Consequences, and Reform; Kaufmann—Kraay—Zoido-Lobatón: Governance Matters

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. None of them touches the monetary policy autonomy of the central bank; all three concern the publicity of operational management.

3.1 A statement at statutory level: the operating expenditure of the central bank is public in the public interest (in the autumn session)

MIAK asks that Parliament, by harmonising the central bank act and the Freedom of Information Act, state that the general publication obligation applies to the MNB’s non-monetary, operating expenditure — procurement, ordering of services, use of real estate, grants, consultancy assignments — irrespective of whether the institution has the status of a budgetary body. The essence of the proposal is not the statement of the principle, because the Constitutional Court already did that in 2016, but the clarification of the statutory text: as long as the content of the obligation is a question of interpretation, every data request restarts the same dispute. The statutory text also has to contain a narrow, itemised list of exceptions, because full publicity is not justified in this area: contracts connected with the management of foreign exchange reserves, with market operations and with the security elements of banknote production genuinely cannot be published immediately, for business or security reasons. The solution for these is deferred publication: the data of the contract automatically becomes public after a period fixed in advance — for instance two years — and the central bank signals the fact of the deferral, with reasons, at the moment of the deferral. This structure is the common logic of the A1 public money dashboard and the G19 radical transparency programme points: not everything immediately, but everything according to a rule known in advance.

3.2 Connecting central bank contracts into the common public money register (within 12 months)

The second proposal concerns technical implementation, and in MIAK’s view this is the more important one. Publicity in itself is not enough if the data appears on a separate website, in an unsearchable form, with a different data structure. MIAK asks that the data of the MNB’s contracts above five million forints be entered into the same register, with the same fields and in the same machine-readable format, as the contracts of the other bodies performing public tasks — that is, into the public money dashboard under A1. It is the uniform data structure that makes the data usable: without it contract publicity means only that the document is in principle knowable, while nobody can demonstrate the overlaps between contracting partners, the repeated assignments or the divergence from the market price. The connection also extends the scope of the A2 public procurement transparency programme point to an institution that has so far remained outside it. MIAK proposes a deadline of twelve months because the data reporting interface already exists: there is no need to build a new system, merely to connect a further data owner to an existing one.

3.3 A cost-bearing rule for successful data request lawsuits (in the next package of procedural law amendments)

The third proposal concerns the price of enforcing rights. The present case too shows the structural problem well: after a refusal, the organisation or citizen requesting the data is forced into an administrative procedure and then typically into litigation, and has to bear the cost and the time of this in advance — while on the side of the institution withholding the data the refusal carries no substantive price. MIAK asks that the legislator fix that if the data requester wins the lawsuit for the release of public-interest data, their full procedural costs — including the fee of legal representation, subject to an itemised upper limit — should fall on the data controller, and that the authority should adjudicate such cases out of turn. This is the concrete content of the I6 programme point on reducing the threshold of citizens’ enforcement of rights, applied to a well-delimited group of cases. The argument is not that institutions have to be punished, but that in the present structure withholding is free while enforcement is expensive — and where costs are distributed in this way, the extent of publicity depends not on the law but on the data requester’s wallet.

The three proposals are held together by a single principle: the scope of independence should be stated, not inferred. According to Klitgaard’s formula (see 6.4.1) corruption risk is produced not by bad intent but by structure — monopoly position and discretionary power without accountability — and of these three elements the present proposals strengthen the last, without touching the other two at all. This matters because there is no genuine conflict between protecting central bank autonomy and procurement publicity: one is justified by monetary decisions, the other by the use of public money, and the two concern different subjects.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Transparency Central bank operating expenditure becomes traceable on the same interface and in the same format as that of other bodies performing public tasks The exception list, if it turns out broad, creates a new legal basis for secrecy instead of publicity
Institutional independence Separating monetary decision autonomy and management publicity strengthens the credibility of the central bank If the regulation does not sharply distinguish the two areas, the publicity of market operations may cause real market harm
Enforcement of rights The cost-bearing rule substantially reduces the price of a data request and speeds up administrative handling Out-of-turn adjudication draws administrative capacity away from other groups of cases
Administrative practice The uniform data structure rules out similar future disputes from the outset, because no interpretative latitude remains Extending the data reporting obligation temporarily imposes significant administrative work on the institution concerned

The main question for deliberation is the breadth of the exception list. If it is too narrow, genuinely market-sensitive information may come out prematurely — in the area of foreign exchange reserve management or banknote security this could cause concrete, quantifiable harm. If it is too broad, the list itself becomes an instrument of secrecy, and the amendment achieves no more than placing the existing dispute on a new statutory footing. In MIAK’s view the solution lies not in the length of the list but in its structure: the exception should always be a deferral, not an exemption, so every withheld piece of data should have a date of becoming public fixed in advance, and the fact of the deferral — without the content of the data — should have to be signalled at the moment of withholding. The proposal tips over to the risk side if the statutory statement is made but the connection to the uniform register is not: in that case the data becomes formally public while remaining practically unusable. There is a measure of ideology-free conduct here too: the rule now proposed would apply to whoever leads the MNB at any given time, regardless of who appointed them — and that is precisely what gives it its credibility.

Part V — Measurability and summary

5.1 What is worth following? (proposed KPIs)

The performance indicators below (KPIs, Key Performance Indicators) will show in 12 and 24 months whether the dispute has produced a working rule. These are proposed benchmarks, not government commitments.

  • The turnaround time of the administrative procedure: in how many days a substantive decision is born in the present NAIH submission — this is the simplest indicator of the real speed of enforcing rights.
  • Whether the statutory statement is made: whether a provision fixing the publicity of central bank operating expenditure gets into legislation adopted in the autumn session, and how extensive the list of exceptions is.
  • The coverage of the uniform register: how many central bank contracts appear in the common, machine-readable public money register by mid-2027, and what percentage this is of all contracts above five million forints.
  • The share of deferred publications: in how many cases, relative to all published contracts, the institution makes use of deferral, and after how much time on average these become public.
  • The number of refused data requests: the central bank’s annual data request statistics — it is worth following what percentage of the requests received it complies with substantively and what percentage it refuses.

5.2 Summary

MIAK’s request to the decision-maker is three steps. In the autumn session the law should state that the publication obligation applies to the operating expenditure of the central bank irrespective of organisational classification, with a narrow list of exceptions built on deferral. Within twelve months central bank contracts above five million forints should be entered in the uniform, machine-readable public money register. And the next procedural package should settle that the costs of a successful data requester be borne by the data controller. And to the public, one request as to outlook: this case is not worth judging on the “is the MNB good or bad” axis. The question is how far the autonomy of an institution extends — and this question will arise in exactly the same way under the next central bank leadership.

Two MIAK foundational values are directly in play here. Transparency, because the knowability of the use of public money is not a favour depending on the goodwill of the institution but the natural consequence of the character of public money — this is what the Constitutional Court said ten years ago as well. And accountability, because autonomy becomes defensible precisely when the operation of the autonomous institution is verifiable: in the long run monetary decision-making freedom is protected from political attack not by secrecy but by demonstrably clean management.

Part VI — Justifications and further sources

6.1 The framing of the press, spectrum by spectrum

The liberal-left band framed the matter as legal-historical continuity, and this framing supplied the most factual data. 444.hu led through the course of the procedure in a detailed piece — the submission of the data request, the extension of the deadline, the refusal of 13 August — and expressly tied the news to the 2016 “lost its character as public money” amendment, recalling the literal finding of Constitutional Court Decision 8/2016. The paper also highlighted that the present central bank governor voted for that amendment as a member of Parliament. HVG carried the same thing with a shorter but sharper headline: according to the central bank they should not have to publish everything about what they spend public money on. The logic of the framing rests on personal continuity: not a new practice, but the continuation of an old one.

The economic band carried the same story as a question of procedural law. Portfolio remained fact-reporting: it recorded the subject of the data request, the central bank’s argument — that it is not a budgetary body — Transparency’s counterargument invoking the 2016 decision, and the procedural objection that after the extension of the deadline the refusal was no longer lawful. This framing is the most neutral: it examines not the intent of the actors but the lawfulness of the steps, and precisely for that reason it is the most usable for policy analysis.

The general public-affairs band highlighted the turn to legal remedy: ATV carried the approach to the authority as the deployment of the rights defenders’ “final trump card”, while Telex placed the fact of turning to the NAIH at the centre of its headline. The conservative band did not carry the topic on this day: Magyar Nemzet’s central bank piece was about an MNB study proposing a rethinking of EU cohesion policy. This absence is itself part of the framing — the question of central bank data publicity did not come into the focus of the conservative band on this day, while the same band is regularly present in other cases of institutional transparency.

6.2 Facts and data

Data Value Source
Submission of the data request early July 2026, Transparency International Hungary 444.hu
The scope of the data requested the MNB’s contracts for consideration and free-of-charge contracts after 1 January 2025 444.hu, Portfolio
The contract value threshold concerned 5 million forints Freedom of Information Act general publication list, point III/4
The extension of the deadline 24 July 2026, +15 days 444.hu
Date of the refusal 13 August 2026 444.hu, Portfolio
Ground of the refusal the MNB is not a budgetary body 444.hu, Portfolio
Turning to the authority 21 August 2026, NAIH Telex, Portfolio, ATV
The Constitutional Court decision invoked Decision 8/2016. (IV. 6.) AB 444.hu
Date of the Constitutional Court decision 6 April 2016 444.hu
The proposition of the Constitutional Court decision the MNB performs a public task and manages exclusively public money 444.hu
The scope of data currently published by the central bank grants of at least 5 million forints 444.hu
Hungary’s governance quality indicator (control of corruption) −0.17 (2024) World Bank, Worldwide Governance Indicators

Two rows of the table together yield the non-obvious conclusion of this post. In 2016 the Constitutional Court stated that the MNB manages exclusively public money; yet in 2026 the central bank can successfully invoke the fact that it is not a budgetary body. In legal technique the two do not contradict each other — the Constitutional Court speaks of the character of public money, the central bank of organisational classification — and it is precisely this gap that the statutory text under proposal 3.1 would close. As long as the statement of principle and the addressee of the obligation live in two different statutory layers, every data request restarts the same dispute.

6.3 Policy dimensions

  • Transparency and anti-corruption policy (programme points) — extending the public money register and the uniform structure of procurement data: A1, A2, A5;
  • Economy (programme points) — the publicity of economic decision-making as an independent institutional requirement: G19;
  • Justice (programme points) — reducing the cost of enforcing data request rights and legislative impact assessment: I6, I3;
  • Public administration and e-government (background material) — the technical conditions of the uniform, machine-readable data reporting format.

6.4 Literature in detail

6.4.1 Robert Klitgaard: Controlling Corruption

The central idea of Klitgaard’s book is a simple, deliberately formula-like summary of the fact that corruption risk is produced not by individual character but by the structure of decision-making. According to the proposition, illicit gain proliferates where the acting agent is in a monopoly position vis-à-vis their client, possesses wide discretionary freedom, and where accountability towards the principal is weak.

“Illicit behavior flourishes when agents have monopoly power over clients, when agents have great discretion, and when accountability of agents to the principal is weak. A convenient equation: CORRUPTION = MONOPOLY + DISCRETION − ACCOUNTABILITY.”

Klitgaard emphasises in the same place that rules in themselves are neither good nor bad from the point of view of corruption: the same rule may create rent and widen discretionary space, but it may also narrow it or make accountability easier. In the present case this distinction is precisely applicable. The MNB’s monopoly position and discretionary power in procurement are necessary and not in themselves objectionable — a central bank obviously has to decide on its own operation itself. The third factor, however, is a matter of regulatory choice: if the data of the contracts is not public, then accountability is weak, and the result of the formula worsens without anyone doing anything wrong. This argument explains why MIAK’s proposals are directed at publicity and not at decision-making freedom.

📖 Source: Robert Klitgaard: Controlling Corruption

6.4.2 Susan Rose-Ackerman: Corruption and Government — Causes, Consequences, and Reform

Rose-Ackerman’s volume discusses in a separate chapter the dilemma that is the core of the present case: how an institution can be at once protected from political influence and held to account before the public. According to her analysis regulatory discretion can never be entirely eliminated, and therefore protection is provided not by secrecy but by procedural guarantees.

“Regulators must therefore be protected from improper influence: by a transparent appointment procedure, by limits on conflicts of interest, and by a security of tenure that insulates them from political pressure.”

Rose-Ackerman adds that the role of hierarchical supervision shifts: the superior level has to measure not conduct but results, and measuring results requires data. Applied to the Hungarian situation, this line of thought says two things. On the one hand the protective instruments of central bank independence — the fixed term of mandate, irremovability, the prohibition of political instruction — all remain even if procurement data is public; publicity does not conflict with them. On the other hand publicity makes possible precisely what Rose-Ackerman proposes: results-based accountability instead of conduct-based supervision. In this framework data publicity is not a limit on independence but the price at which independence is sustainable in the long run.

📖 Source: Susan Rose-Ackerman: Corruption and Government — Causes, Consequences, and Reform

6.4.3 Kaufmann—Kraay—Zoido-Lobatón: Governance Matters

The work of the three authors is methodological in character: it shows that the quality of governance is not intangible but can be described by aggregate indicators assembled from several sources. The volume separates six basic concepts, and the first of these is voice and accountability.

“We construct six aggregate indicators corresponding to six basic governance concepts: voice and accountability, political instability and violence, government effectiveness, regulatory burden, rule of law and corruption.”

The authors’ main conclusion is that these indicators are connected with development outcomes — governance is therefore not a value for its own sake but has measurable economic consequences. Applied to the Hungarian situation, this gives the conceptual basis of MIAK’s proposed performance indicators. If data publicity is a measurable dimension, then the share of refused data requests, the turnaround time of the administrative procedure and the coverage of the register are not supplementary administrative statistics but indicators of the quality of governance itself — and are therefore worth publishing annually, by the same method.

📖 Source: Kaufmann—Kraay—Zoido-Lobatón: Governance Matters

6.5 International comparison

The publicity of central bank management is not a Hungarian peculiarity, and international practice points in precisely the opposite direction from the present domestic interpretation. The European Central Bank makes public its procurement procedures, its contracting partners and its framework contracts, and also publishes the expenditure items of its management annually; publicity does not affect the confidentiality of the preparation of monetary policy decisions, which is protected by separate exception rules defined by subject. The European Court of Auditors regularly examines the operational management of EU central banking and supervisory bodies and publishes its findings in a public report.

The most instructive parallel, however, is British practice: the Bank of England publishes the minutes of its monetary policy meetings with a fixed, pre-announced delay, while it publishes its operational and procurement data in the same order as other public institutions. This duality is exactly the structure that proposal 3.1 describes as deferred publication: confidentiality is not a perpetual property of a type of data but a question of time. The critical literature on the transparency practice of the international financial institutions — following the work of Joseph E. Stiglitz — confirms the same connection: the legitimacy of technocratic autonomy stands on the operation of the institution being verifiable after the fact.

Transparency and anti-corruption policy

  • A1 — Public money dashboard
  • A2 — Public procurement transparency
  • A5 — Whistleblower reporting system
  • A10 — Independent Corruption Investigation Office

Economy

  • G19 — Radical transparency in economic decision-making

Justice

  • I3 — Legislative impact assessment
  • I6 — Reducing the threshold of citizens’ enforcement of rights

Proposed new programme point: Deferred publication as a uniform exception technique — for the Transparency and anti-corruption policy policy area: the withholding of public-interest data should in every case be a deferral tied to a date of becoming public fixed in advance, not an exemption of indefinite duration.

6.7 List of sources

Press sources (MIAK press monitor, 22 August 2026 — topic 2):

Knowledge base references (literature):

  • 📖 Robert Klitgaard: Controlling Corruption
  • 📖 Susan Rose-Ackerman: Corruption and Government — Causes, Consequences, and Reform
  • 📖 Kaufmann—Kraay—Zoido-Lobatón: Governance Matters

MIAK internal materials:

  • MIAK policy area: Transparency and anti-corruption policy (programme points and background material; programme point ID: A1)
  • MIAK policy area: Economy (programme points; programme point ID: G19)
  • MIAK policy area: Justice (programme points; programme point ID: I6)
  • MIAK policy area: Public administration and e-government (background material)
  • MIAK press monitor, 22 August 2026 — topic 2, score: 90/100

Supplementary public data sources:

  • National Authority for Data Protection and Freedom of Information — annual reports and decision database
  • Transparency International Hungary — reports and data request documentation
  • World Bank, Worldwide Governance Indicators — governance quality time series

Generation metadata