Part I — Situation overview

On 27 and 28 August 2026 four mutually independent official proceedings became known. On 23 August the National Bureau of Investigation of the Rapid Response Police launched criminal proceedings against an unknown perpetrator on suspicion of misappropriation. According to the reports of Portfolio and HVG, MÁV Személyszállítási Zrt. waived in substantial part the debts of several municipalities arising from local passenger transport — an item of some 2 billion forints for Kecskemét, roughly 1 billion for Salgótarján, close to 600 million for Nyíregyháza and about 500 million forints for Szombathely. On 26 August the same investigating authority ordered proceedings over the 79.6 billion forint capital increase carried out at Paks II Atomerőmű Zrt. in December 2025. In a separate case, following a criminal complaint by Transparency International, the Budapest Police Headquarters is examining whether Dolomit Kft., the company of Győző Orbán, supplied stone at an overpriced rate for the widening of the M1 motorway. Finally, on 28 August the Ministry of the Interior filed a criminal complaint with the Criminal Directorate General of the National Tax and Customs Administration (NAV) over the 24.947 billion forints provided in 2016 and 2017 for the Fradiváros project of Ferencvárosi Torna Club.

In all four cases the antecedents go back years, and in all four the same thing recurs: the legal form was in place, the written professional statement of reasons for the decision was not. In the Fradiváros case the grant instruments were amended five times per phase between 2020 and 2026, and a government decision of late 2022 extended the purpose of use to operating expenditure. In the case of the waived local transport debts the ministry still ordered collection in 2024, and a year later showed leniency.

MIAK’s reading: this is not four scandals but one structural fault appearing four times. Where the state takes a decision on assets — waives a debt, increases capital, purchases from a related party, modifies the purpose of a grant — there is today no obligatory, uniform statement of reasons that can be read afterwards. As long as this remains so, responsibility can only be a criminal-law question, whereas it is primarily a procedural question. There is an important delimitation which MIAK maintains for the sake of its own credibility as well: the ordering of an investigation is not the establishment of guilt, a criminal complaint is not a proceeding, and the presumption of innocence applies without exception to everyone concerned.

Part II — Foundations in the literature

Before we turn to MIAK’s proposals it is worth setting down the interpretive framework in which the shared feature of the four cases becomes visible. According to the formula of Robert Klitgaard (former professor at the Harvard Kennedy School, author of the classic handbook of anti-corruption policy), corruption flourishes where monopoly position and discretionary power meet the absence of accountability; the volume unfolds precisely the point that a rule is not good or bad in itself, but according to whether it widens or narrows actual freedom of discretion. The concept of János Kornai (economist of Hungarian origin, author of the theory of the soft budget constraint, professor at Harvard between 1986 and 2002) explains why these cases cluster precisely at state-owned companies: if losses are regularly paid off, the behaviour of the company becomes detached from its own solvency, and the place of market bargaining is taken by redistributive bargaining conducted with the authorities. And the criminal law in force provides the conceptual precision which daily press language blurs: the statutory definition of misappropriation is not appropriation but the breach of the duty attaching to the management of assets entrusted to one. The detailed treatment of the literature — author by author, with quotations — can be found in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. All three relate to the future, none of them touches the ongoing proceedings, and none of them requires the setting up of a new institution.

3.1 An obligatory written statement of reasons for every discretionary act concerning state assets (together with the 2027 budget act)

Parliament should prescribe that the waiver of a claim of a company in state or municipal ownership, a capital increase, a state guarantee, and a contract concluded with a related party above a specified value threshold — in our proposal 500 million forints — are valid only if a written statement of reasons is prepared for the decision according to a set of criteria published in advance. The statement of reasons has to contain at least four elements: what the objective to be achieved is, what alternative was considered, how large the estimated effect of the decision on the owner’s assets is, and which organisational unit is responsible for preparing the decision. In Klitgaard’s C = M + D − A framework (see 6.4.1) this is precisely the narrowing of D, that is, of discretionary power — not the removal of the right to decide, but the obligation to give reasons for the decision. The proposal is the asset-side of the G19 radical transparency programme point, and it does not infringe business secrets: where the statement of reasons genuinely contains a business secret, the secret part can be sealed, but the fact of the decision, its amount and the responsible body cannot.

The summary of the statement of reasons under 3.1 should be put within fifteen days onto the public money page proposed by MIAK (A1), in a machine-readable, queryable format. An obligatory field of the record should be the related-party character: if the ownership circle of the contracting partner touches at any point the circle of the decision-maker or of the exerciser of ownership rights, the record has to indicate this. This is the point at which the present system is most conspicuously deficient: in the matter of the stone deliveries connected to the widening of the M1, the press tried for years to reconstruct through chains of subcontractors who bought from whom and for how much, because the main contractor’s contracts were public while the circle of subcontractors was not. The flag in itself prohibits nothing — a contract concluded with a related party can be entirely lawful and economical — it merely makes it visible. The solution is technically ready: the A2 public procurement anomaly detector would run on this same database, and would already flag recurring winner–contracting authority pairs today.

3.3 An annual return examination on the capital injected into state companies (first report by the end of 2027)

For every state capital injection above 10 billion forints a subsequent examination should be prepared each year, comparing the expected result recorded at the time of the injection with the actual one. The result of the examination should be public and the methodology fixed in advance, so that the report is not a retrospective judgement but a comparable time series. This is the direct application of MIAK’s G20 programme point, that is, of the Drucker audit — the method of Peter F. Drucker, the Austrian-American management thinker, which prescribes the subsequent comparison of the expected and the actual result. The Paks II case is instructive because here the tension between the physical state of the investment and the order of magnitude of the outlay is publicly visible, yet there is no annually repeated statement of identical methodology from which this could have been followed for years. The audits of the State Audit Office (ÁSZ) are not sufficient for this: the ÁSZ is a supervisory, not a binding body, its report makes findings, but regular, comparable publication is the task of the exerciser of ownership rights.

The three proposals run out into a single principle: a decision on state assets is legitimate if it can be reconstructed afterwards. Not because the decision-maker is suspect from the outset, but because reconstructability is the only instrument which also protects the good-faith decision-maker. For in the present system defence too is impossible: where there is no written deliberation, neither responsibility nor the absence of responsibility can be proved.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Economy The return on capital injected into state companies becomes comparable; the soft budget constraint tightens, because paying off a loss leaves a trace The obligation to give reasons may slow down genuinely urgent ownership decisions (liquidity rescue, wage guarantee)
Society Decisions on public assets can for the first time be followed from a single interface, not from press investigations Raw data without context can be misinterpreted; a flagged related-party relationship may automatically appear to be a suspicion
Public administration The work of the decision-preparing apparatus becomes documented and therefore defensible The spread of formal templates for statements of reasons, which fill the box but contain no substantive deliberation

The most serious question for deliberation is the relation between speed and documentation. The liquidity crisis of a state company can escalate within hours, and it is an unrealistic expectation that a full impact assessment should precede the decision at such a time. The solution is not the lifting of the obligation but subsequent completion with a strict deadline: the emergency decision may be taken without a statement of reasons, but if the statement of reasons is not prepared within thirty days and the exerciser of ownership rights does not publish it, the decision falls under automatic review. This is the same structure which MIAK proposes for the I3 legislative impact assessment programme point as well.

The second risk is templating. If the statement of reasons becomes a sentence selected from a drop-down menu, the rule is formally fulfilled but not in substance. There is only one effective instrument against this: publicity. If the statements of reasons are available in machine-readable form, recurring templates can be detected statistically at once, and the yardstick is not an authority but public comparison.

Part V — Measurability and summary

5.1 What is worth following? (proposed KPIs)

On the basis of the performance indicators (KPIs, Key Performance Indicators) below it will be possible in 12–24 months to judge whether the proposal works. These are proposed indicators, not government undertakings.

  • Coverage by statements of reasons: for what percentage of discretionary decisions on state assets above 500 million forints a written, published statement of reasons was prepared. Proposed target: above 90 per cent by the end of 2028.
  • Publication delay: the median number of days elapsed between the decision and its publication. Proposed target: under 15 days.
  • Completeness of the related-party flag: for what percentage of published asset transactions the related-party field was filled in substantively. Proposed target: full completeness.
  • Number of return reports: how many annual statements of identical methodology were prepared on state capital injections above 10 billion forints. Proposed target: coverage of all the injections concerned from 2028.

5.2 Summary

MIAK’s request in a single sentence: the Government and Parliament should not spend the autumn commenting on the four cases; instead the Government should submit, together with the 2027 budget act, and Parliament should adopt, the obligation to give reasons for and to publish discretionary decisions on state assets. This is the only step which brings a benefit irrespective of the outcome of the present proceedings: if the investigations end in discontinuation, the rule which in the coming decade prevents similar situations still remains; if they end in indictment, the rule will be no less necessary for that.

Two MIAK foundational values move together here. Transparency, because the decision trail is not a punishment but the practical form of the duty of account of organisations managing public money, a duty which can also be derived from the Fundamental Law. And accountability, because holding to account loses its meaning if it can take place only retrospectively, in individual cases, by the instruments of criminal law. In a state under the rule of law the good system is the one in which the question of responsibility mostly does not even arise, because every step of the decision is documented — and precisely for this reason MIAK expresses no opinion on the proceedings now under way, only on what was missing when the decisions were taken.

Part VI — Justifications and further sources

6.1 The framing of the press, spectrum by spectrum

The Hungarian press did not treat the four cases as a single story, and the greatest difference between the segments lies precisely in which case they place in which frame.

The left-liberal segment presented the cases as a coherent process of holding to account. Telex reported on all three misappropriation proceedings in separate articles, while HVG tied the waived transport debts expressly to the campaign period and referred back to its own document-based investigation in May. In the M1 stone delivery case 444.hu emphasised the long investigative antecedent: that Dolomit Kft. remained for years in a subcontractor position, and so the chain of contracts could not be reconstructed from the outside.

The economic segment chose a narrower but more disciplined frame. In its account of the transport debts Portfolio indicated in a separate sentence that the claim concerning political motivation is an interpretation, and that there is no final official or judicial decision on it — this kind of self-restraint is rare in the daily news flow, and it is precisely the distinction which MIAK also maintains.

The public affairs segment focused on the practical consequences of the cases: in the Fradiváros case 24.hu brought forward the concrete conditions of the repayment obligation — a lump-sum payment within thirty days, payment in twelve monthly instalments or a subsequent audit — and looked in a separate article at what had been fulfilled of the seven official obligations which the ministry imposed on MÁV in 2024.

The conservative segment picked out essentially one of the four cases, the Fradiváros case, and even that in a fact-reporting register: Mandiner gave an account of the communiqué of the Ministry of the Interior and of Gábor Kubatov’s reply, without commentary. The proceedings connected with Paks II and the waived debts did not come into focus in this segment on that day. This in itself is information: certain elements of the process of holding to account reach readers today not on a shared factual basis but through a selection which differs from segment to segment.

6.2 Facts and data

Case Public money concerned The nature of the proceedings Date
Waived local transport debts (Kecskemét, Salgótarján, Nyíregyháza, Szombathely) more than 4 billion HUF criminal proceedings on suspicion of misappropriation, against an unknown perpetrator 23 August 2026
Capital increase at Paks II Atomerőmű Zrt. 79.6 billion HUF investigation on suspicion of misappropriation causing particularly substantial pecuniary disadvantage 26 August 2026
Stone deliveries for the widening of the M1 on the basis of the framework contract, 4.8 billion HUF of revenue in 2025 investigation on suspicion of misappropriation, following the criminal complaint of a civil organisation 27 August 2026
Support for the Fradiváros project 24.947 billion HUF ministerial criminal complaint, in parallel with the ordering of the repayment of the support and its interest 28 August 2026

The figures of the Fradiváros case also show the structure of grant management. Of the 10 billion forints of the first phase, according to the final report more than 8.5 billion forints appear as material and operating expenditure and some 1.44 billion forints as investment. Of the 14.9 billion forints of the second phase more than 3.1 billion is operating expenditure, while 11.2 billion went to FTC Labdarúgó Zrt. The grant instruments were amended five times per phase between 14 October 2020 and 18 February 2026, and the extension of the purpose of use was made possible by a government decision of 28 December 2022.

The antecedent of the Paks II case is that in September 2025 the Court of Justice of the European Union annulled the 2017 Commission decision which had approved the Hungarian state aid; the financial performance of the 79.6 billion forint capital increase took place after this, by 27 December 2025.

As institutional context it is worth recording: according to the data of the World Bank’s Worldwide Governance Indicators, in 2024 Hungary reached a value of −0.17 on the control of corruption indicator, while government effectiveness stood at +0.42 and the rule of law at +0.35. The divergence of the three values traces precisely the pattern this entry is about: the apparatus works, the rules exist, the holding to account of decisions based on deliberation is weak.

6.3 Policy dimensions

  • Transparency and anti-corruption policy (programme points) — the public money dashboard (programme point ID: A1) and the public procurement anomaly detector (programme point ID: A2) would provide the technical infrastructure of the proposal; for the identification of related parties the machine readability of asset declarations (programme point ID: A3) is also necessary.
  • Economy (programme points) — the statement of reasons for decisions is the asset-side application of the radical transparency programme point (programme point ID: G19), and the return examination that of the Drucker audit (programme point ID: G20).
  • Transport and infrastructure (programme points) — the settlement of local transport debts and the data-based grounding of the order of railway investments (programme point ID: KO4) are two sides of the same question: in both the stake is whether the professional criterion or ad hoc deliberation decides.
  • Justice (programme points) — the subsequent completion of the statement of reasons for emergency decisions follows the logic of the legislative impact assessment system (programme point ID: I3).

6.4 Literature in detail

6.4.1 Robert Klitgaard: Controlling Corruption

Klitgaard’s proposition is well known — corruption = monopoly + discretion − accountability — but its less frequently cited part is precisely the decisive one from the point of view of the present proposal. The volume treats separately the apparent contradiction that regulation can be at once the instrument and the remedy of corruption, and traces the distinction back to what the rule does to actual freedom of discretion:

“But a rule may be used to reduce discretion, thereby reducing corruption: for example, a tax officer may be given no discretion about allowable deductions, or an admissions committee may be told to select students on the grounds of test scores. A rule may also be used to make accountability easier, which should help reduce corruption.”

Klitgaard adds: rules in themselves are neither good nor bad from the point of view of corruption — they may create or eliminate rents, widen or narrow actual discretion, make accountability easier or harder. This is precisely why the four Hungarian cases can be placed on a single axis. The fivefold amendment of the grant instrument, the waiver of the claim, the capital increase and the subcontracted stone delivery are problematic not because they were unregulated, but because in every case the rule left deliberation open and nowhere tied it to an obligation to give reasons. MIAK’s proposal 3.1 closes precisely this gap: it does not take away the right to decide, it makes accountability possible.

📖 Source: Robert Klitgaard: Controlling Corruption

6.4.2 János Kornai: A hiány (Economics of Shortage)

Kornai introduces the concept of the soft budget constraint not as a moral but as a behavioural category: the question is not whether the company is loss-making, but whether the consequence of the loss is a matter of life and death for it. Where it is not, the conduct of the company becomes detached from its own solvency:

“The survival of the firm depends not only on whether it is able to cover, over a sustained period, the expenditure on the purchase of its inputs from the revenue of the sale of its output. Even if the latter exceeds the former over a sustained period, this may be offset by tax concessions, state grants, soft credit and so on.”

Kornai also describes the consequence which is the most important from the point of view of today’s cases: administrative dependence takes the place of market dependence, and the place of plan bargaining is taken by redistributive bargaining, in which the company haggles with the authorities over taking less from it and giving more to it. This is exactly the structure in which the waiver of a claim by a state passenger transport company or the capital injection into a state investment company becomes not a business but a bargaining question. MIAK’s proposal 3.3 — the annually repeated return statement of identical methodology — does not prohibit the bargain but puts a price on it: whoever receives capital has to show every year what came of it.

📖 Source: János Kornai: A hiány (1980)

6.4.3 Act C of 2012 — the Criminal Code

Press language often describes the four cases as “stolen public money”, but the suspicion designated by the authorities is more precise and narrower than that. According to the Criminal Code:

“Section 376 (1) Any person who is entrusted with the management of the assets of another and who causes pecuniary disadvantage by breaching the duty arising therefrom commits misappropriation.”

At the centre of the statutory definition, therefore, stands not appropriation but the breach of the duty attaching to the management of the assets entrusted to the person. This difference is substantive in the present cases: the question in every instance is whether the decision-maker acted in accordance with the obligation relating to the assets entrusted to them, not whether the money ended up in their own pocket. From this follows one of MIAK’s most important claims: if the content of the duty were determined by a written set of criteria fixed in advance, then both its breach and its observance could be proved. In the present state of affairs, however, both are questions of evidence lasting years. The law in force does in any case treat the graver cases separately: the penalty for misappropriation causing particularly substantial pecuniary disadvantage is imprisonment from five to ten years.

📖 Source: Act C of 2012 (Criminal Code of Hungary), Section 376

6.5 International comparison

There are two established European patterns for enforcing the decision trail, and both show that the effect depends not on the severity of the supervision but on the automatism of publication.

In Slovakia, since 2011 every state contract enters into force only if it has been published in the central contract register. The subtlety of the solution lies in the fact that publication is not a subsequent obligation whose omission is to be sanctioned, but a condition of validity — so no separate supervisory apparatus is needed. After the introduction of the measure, public procurement prices moderated in the transparent categories.

In Ukraine the ProZorro system introduced in 2016 brought the share of single-bid procedures down from around 40 per cent to 18 per cent. The lesson for the Hungarian situation is that the publicity of data is not enough in itself: ProZorro’s strength came from the fact that a civil analytical community was built on the data, which continuously flagged the anomalies.

In Brazil, the experience of the Portal da Transparência, operating since 2004, is that a significant part of the corruption cases uncovered started from civil data analysis. This at the same time marks out the limit of MIAK’s proposal: publication works if the data arrive in a queryable form, not as an image or a scanned document.

Transparency and anti-corruption policy

  • A1 — Public money dashboard
  • A2 — Public procurement transparency
  • A3 — Publicity of asset declarations
  • A8 — Cohesion policy accountability

Economy

  • G19 — Radical transparency in economic decision-making
  • G20 — Economic policy impact assessment system (Drucker audit)
  • G21 — Systematic review of state expenditure

Transport and infrastructure

  • KO4 — Railway development with data-based priorities

Justice

  • I3 — Legislative impact assessment

Proposed new programme point: An obligatory statement of reasons for decisions in state asset acts based on deliberation — for the Transparency and anti-corruption policy area.

6.7 List of sources

Press sources (MIAK press monitor, 29 August 2026 — topic 2):

Knowledge base references (literature):

  • 📖 Robert Klitgaard: Controlling Corruption
  • 📖 János Kornai: A hiány (1980)
  • 📖 Act C of 2012 (Criminal Code of Hungary)

MIAK internal materials:

  • MIAK policy area: Transparency and anti-corruption policy (programme points; programme point ID: A1, A2, A3, A8)
  • MIAK policy area: Economy (programme points; programme point ID: G19, G20, G21)
  • MIAK policy area: Transport and infrastructure (programme points; programme point ID: KO4)
  • MIAK policy area: Justice (programme points; programme point ID: I3)
  • MIAK press monitor, 29 August 2026 — topic 2, score: 91/100

Supplementary public data sources:

  • World Bank — Worldwide Governance Indicators, Hungarian values for 2024
  • Electronic financial statements portal (e-beszamolo.im.gov.hu) — the annual reports of Paks II Atomerőmű Zrt. and the MÁV group
  • Court of Justice of the European Union — the judgment annulling the Paks II state aid decision (2025)

Generation metadata