Part I — Situation overview

On Monday, 28 September 2026, the National Asset Recovery and Asset Protection Office (NVVH) makes public the four cases it is the first to take over from other authorities. Today’s date was named on 18 September, at HVG’s Címlapsztori event, by Anna Unger, the office’s president elected by the National Assembly at the end of August. At the time she said only that these were “four very good” cases. According to the MTI summary carried by Telex, Népszava and ATV, the office may take over only cases in which at least one criminal complaint has already been filed, and the selection is decided by Katalin Tasnády, the vice-president responsible for investigations. The office’s criminal-procedure powers took effect on Sunday, 27 September. According to the information published on mnvvh.hu, the bodies from which the cases are taken over are also notified before the announcement. According to the same information, the office’s aim is not solely to examine criminal liability, but to trace, secure and recover unlawfully acquired public assets. At the time of writing, on Monday morning, the names of the four cases are not yet known. This analysis therefore does not assess the cases, but sets out the standard against which today’s announcement can be measured.

The announcement is surrounded by a strong political framing. On Sunday evening Prime Minister Péter Magyar wrote that within the framework of Operation Purgatory “the landing begins” on Monday, and that the National Assembly “hands over two corrupt politicians to the investigators”. In legal terms, on the same day, at an extraordinary sitting, the National Assembly may decide, on the motion of the deputy Prosecutor General, on suspending the parliamentary immunity of two former ministers and of the Prime Minister himself. Suspending immunity makes proceedings possible; it is not a finding of guilt. Viktor Orbán, chairman of Fidesz, called the office an ÁVH (the political police of the Rákosi era) on Hír TV, and said that the names of its staff would be made public. Anna Unger replied that if the staff were intimidated, the office would take legal action. On the same morning, 24.hu published data from the Ministry of Foreign Affairs: in the previous parliamentary term, 286.9 million forints were paid for private jets on which former minister Péter Szijjártó ultimately did not travel. The ministry had earlier refused to release the same data, citing foreign relations. On 24 September HVG compiled into a table 33 cases in which a criminal complaint had been filed.

MIAK has already published several analyses on the setting-up of the NVVH, the selection of its head and its first-year indicators (The nomination of Anna Unger, The NVVH gets going, case-status register). What is new today is that the institution is appearing before the public with concrete cases for the first time. MIAK’s reading: the office’s credibility today depends less on which four cases it chooses than on how it presents them — with facts and figures, or with names and epithets. The form of the first announcement will be a precedent for every subsequent one, and under every subsequent government.

Part II — Foundations in the literature

Three classical and empirical sources provide the interpretative framework. Montesquieu (French legal philosopher, classical author of the separation of powers) writes in The Spirit of the Laws that under moderate government confiscation renders property uncertain, and must therefore be limited to a narrow range of crimes. He also warns that if the power itself enjoys the benefit of confiscation, it is both judge and party. Robert Klitgaard (American economist, author of one of the foundational works on anti-corruption institutional design) describes the “big fish” principle in Controlling Corruption: the credibility of a new anti-corruption body is established by its first high-ranking cases, and the public measures effectiveness by the rank of those prosecuted. Corruption and Government by Susan Rose-Ackerman (professor of law and political science at Yale University, researcher of the political economy of corruption) uses the example of the Hong Kong anti-corruption commission to show how staff were protected by institutional rules, and from what the growth of credibility could be demonstrated. The detailed discussion of the literature — by author, with quotations — can be found in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable steps. The first sets the form of today’s announcement, the second the procedural guarantees of case takeovers, and the third the yardstick of success.

3.1 A uniform case card without names (from today’s announcement; publication of the selection rules within 30 days)

MIAK proposes that the NVVH present all four cases on short case cards of identical structure, and keep this format for every later announcement. The case card should contain six elements: (1) the subject of the case described from the public-money side, that is, which contract, subsidy or asset transfer, in which period, under the commitment of which state body; (2) the date of the criminal complaint and the authority from which the office is taking over the case; (3) the stage of the proceedings, for example whether there is already a suspect, but without names; (4) the estimated amount of public assets affected, with the source of the estimate; (5) the amount of assets already secured, that is, frozen or seized, if any; (6) the reason for the selection according to criteria fixed in advance. Such criteria include the size of the damage, whether the assets can be secured, the risk of limitation periods expiring, and the exemplary value of the case. The latter means that the case also helps uncover similar, recurring schemes. The case card should not contain the name of any natural person whom the authorities have not made a suspect, and it should not characterise suspects with epithets either. Under Section 1 of Act XC of 2017 on Criminal Procedure (Be.), no one may be considered guilty until this has been established by a final court decision. The Act also sets limits on informing the public. The voice of the announcement should be that of the office, not of the government: the “landing” is political speech; the office’s language is the case card. According to Klitgaard’s principle (see 6.4.2), the first cases prove the office’s seriousness. Precisely for this reason the reason for the selection must also be published; otherwise choosing the “big fish” looks like a political decision. This is also how the distinction between the four cases and the 33 cases compiled by HVG becomes understandable. The proposal builds on the data logic of the A10 independent investigation office and the A1 public-money dashboard.

3.2 Reasoned, reviewable case takeovers and institutional protection of staff (simultaneously with the first takeovers)

The NVVH joins cases that are already in progress. This is the most vulnerable point: if a procedural error occurs around the takeover, it may jeopardise the whole case at the later court stage. MIAK therefore proposes that every case takeover be the subject of a written, reasoned decision, to be received by the authority that has handled the case so far, the injured state body, and the accused and defence counsel. The decision should record, in a handover report, which documents, seized assets and pending deadlines the office takes over. This rules out two authorities investigating the same case in parallel, or deadlines being lost during the handover. The takeover decision should be reviewable before an independent forum. Index’s opinion piece of 28 September claims that certain decisions of the office are not subject to judicial control. If this is so, MIAK proposes that the next amendment of the law remedy it. Judicial oversight does not weaken the office; it protects its cases from later annulment. MIAK’s position on the staff list is clear: the names of the office’s leaders are already public, and those of the investigators need not be. Accountability comes not from personal exposure but from the traceability of decisions: every procedural act should be carried out by an identifiable person known to the court and the defence. According to Rose-Ackerman’s account (see 6.4.3), the staff of the Hong Kong commission were protected by institutional rules. The proposal transfers the logic of the A6 checks-and-balances and the I4 independence programme points to the new office.

3.3 A baseline value at announcement and a recovery rate as the yardstick of success (from today’s announcement, with quarterly publication)

MIAK proposes that for each of the four cases the office fix a baseline value as of the day of today’s announcement: the estimated damage and the assets already secured. Against this, it should publish five data points per case every quarter: the amount of assets secured, whether charges have been brought, the final judgment, the amount of confiscation ordered or civil-law claim awarded by final decision, and the amount actually returned to the state. The office’s main indicator should be the recovery rate, that is, the ratio of the amount actually returned to the damage estimated at the time of the announcement. The yardstick should not be the number of cases opened or taken over, nor the size of the amounts announced. According to Portfolio’s guest author, criminal lawyer Balázs Gábor Fodor, asset-securing and confiscation statistics are fragmented today, so the effectiveness of asset recovery is practically impossible to verify. The baseline value fills this gap for the first four cases. Anna Unger herself put it, according to Index’s report, that in society the demand for criminal accountability is stronger than the demand for recovering the vanished assets, but for the office the latter is at least as important a task. The indicator must reflect this emphasis. Montesquieu’s warning (see 6.4.1) justifies a further rule. Recovered assets should not increase the office’s budget or its staff’s bonuses, but should go to the central budget or to the injured state body. In this way the office has no financial interest in confiscation, and does not become a judge in its own cause. The proposal applies the KI8 results-based efficiency-measurement programme point to a single office, and the data can appear on the A1 public-money dashboard.

The three proposals are linked by a single principle: every step of an office with strong powers that does not jeopardise the investigation should be verifiable. The case card is addressed to the public, the reasoned takeover to the court, the recovery rate to the budget. Klitgaard and Montesquieu point to the same tension: the new office has to gain credibility quickly, but a spectacular start and a financial interest undermine precisely that credibility if there is no rule alongside them.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Rule of law The uniform, nameless case card and the reasoned case takeover reduce the chance that a court later annuls the cases because of a procedural error A case card that is too terse may look like secrecy, and the gap will be filled by political communication
Public finances The baseline value and the recovery rate show how much of the assets actually come back The final result takes years; in the first quarters the indicator will be close to zero, and this may look like failure
Public administration The handover report rules out parallel investigations and the loss of deadlines The administrative burden of the takeover may slow down the first cases
Staff Institutional protection and the traceability of decisions reduce the effect of personal intimidation If the list of names later leaks nonetheless, the lack of protection also makes recruitment harder
Public life The debate shifts from names and epithets to amounts and procedural stages Political actors may continue to use combative framing regardless of the office’s standard

The main dilemma is time. The recovery rate is the most honest indicator, but also the slowest: charges, first- and second-instance judgments, then enforcement follow, and the amount returned stays low for years. If the office publishes only this, public impatience may push it towards spectacular announcements. If, on the other hand, it publishes only the assets secured, that rewards accelerated freezes that later fail in court. This is why the two figures must be published together and per case: the securing rate shows speed, the recovery rate shows the result, and the gap between the two will itself become a yardstick over time. MIAK’s background material also warns of the transparency paradox: if the public sees many announcements but few results, this produces cynicism, not trust. The second risk is the political framing. The office is not responsible for what the Prime Minister or the opposition says about it. It is, however, responsible for not adopting this language in its own communications, because before the court it is the office’s communications that count.

Part V — Measurability and summary

5.1 What is worth following? (suggested KPIs)

MIAK suggests tracking the following key performance indicators (KPIs):

  • Completeness of the case card: in today’s announcement, for how many of the four cases all six elements appear (subject, transferring authority, procedural stage, estimated damage with source, assets secured, reason for selection). Suggested target: all four, with the selection rules made public within 30 days.
  • Communication without names: how many names of natural persons whom the authorities have not made suspects appear in the office’s own statements. Suggested target: zero.
  • Securing rate after six months: what percentage of the damage estimated at the time of the announcement the assets secured in the four cases represent. Worth tracking quarterly, per case.
  • Recovery rate after 24 months: what percentage of the damage estimated at the time of the announcement the amount actually returned to the state represents, together with the number of final decisions. Worth assessing publicly in September 2028.
  • Review of takeover decisions: how many takeover decisions were challenged, and how many of these the review forum upheld. Worth publishing annually.

5.2 Summary

MIAK asks the NVVH to publish with today’s announcement not only four cases but also a standard. It should present every case in the same format: by public money, amount, procedural stage and reason for selection, without names or characterisations. It should make reasoned, reviewable decisions on case takeovers, and protect its staff with institutional rules, not by keeping the list of names secret or publishing it. It should fix the yardstick of success as of today: how much of the assets come back under final decisions relative to the damage estimated at the time of the announcement, and none of this should add to the office’s own budget. MIAK asks the government and the opposition to measure the office’s work against this standard, not by the volume of its announcements.

Two MIAK foundational values are at play here. Accountability, because an asset recovery office can hold others to account only if its own results are numerically verifiable and its decisions reviewable. Ideology-free approach, because the case card, the takeover rule and the recovery rate are independent of political sides: they could be applied in the same way if a future government’s office were examining contracts concluded under the current government.

Part VI — Reasoning and further sources

6.1 Press framing by spectrum

In the left-liberal band, Telex and Népszava carried the MTI summary, factually, building on the “four very good” cases quote and on the Fidesz demand concerning the list of names. Népszava discussed the announcement together with parliament’s sitting that day. HVG, in its subscriber (HVG360) section, chose a tabular, data-driven framing. It argues that the corruption methods of the previous term were made “from a few templates”, so in a given type of case even a single uncovered instance can serve as the basis for a series of proceedings. 444.hu put the spotlight not on the office but on the assets under scrutiny: it wrote about Ádám Matolcsy’s asset acquisitions in the summer, in parallel with the investigation into the central-bank foundations (the full article is subscriber content). This framing stresses the urgency of securing assets.

In the public-affairs band, 24.hu brought two threads to the fore: the Prime Minister’s “landing” post and Viktor Orbán’s ÁVH comparison. On the same day, this outlet also obtained the Ministry of Foreign Affairs’ data on private-jet cancellation penalties. Index published a detailed table of the penalties, and in an opinion piece Gábor Fodor gave a sharply critical framing: he discussed the office from the angle of the lack of judicial control and a climate of political revenge, citing the UN speech of President of the Republic András Baka, according to which the administration of justice must not become a political weapon. ATV highlighted that the office’s president immediately rejected the intimidation of staff.

In the economic band, Portfolio’s news feed discussed the office’s first announcement together with the immunity cases and the election of the Prosecutor General. In the outlet’s opinion section, criminal lawyer Balázs Gábor Fodor argues that the real test of asset recovery will be not the new institution but measurable results. This framing is closest to MIAK’s point 3.3.

In the conservative band, the office’s first cases did not feature among the highlighted topics on the Monday-morning front pages of Magyar Nemzet and Mandiner. On this day the opposition position appeared mainly through the Fidesz statement quoted by other outlets and Viktor Orbán’s Hír TV interview: it foregrounded making the identity of the staff public and later calling “procedures outside the rule of law” to account. From MIAK’s point of view this framing raises a substantive question but gives the wrong answer to it: the office’s accountability is indeed a question, but it is ensured by the reviewability of decisions, not by the personal exposure of staff.

6.2 Facts and data

Data Value Source
Entry into force of the NVVH’s criminal-procedure powers 27 September 2026 24.hu, Telex, mnvvh.hu, 27–28 September 2026
Day on which the first cases are named 28 September 2026 Anna Unger, HVG Címlapsztori, 18 September 2026 (MTI, Telex, Népszava, ATV)
Number of cases taken over in the first round 4, cases requested from other authorities MTI, 24.hu, 28 September 2026
Condition for takeover at least one criminal complaint in the case Anna Unger (MTI, Telex, ATV)
Leader deciding on selection Katalin Tasnády, vice-president for investigations MTI, Telex, ATV
Election of the president and the vice-president for investigations end of August 2026, National Assembly 24.hu, Index
Cases with criminal complaints in HVG’s compilation 33 HVG, 24 September 2026
Cost of private jets not used, 2022–2026 286.9 million forints, 8 items Ministry of Foreign Affairs data, 24.hu, 28 September 2026
The most expensive item 89 million forints (Cairo trip, September 2024) 24.hu, HVG, Index, 28 September 2026
Chartering of special aircraft in total, 2022–2026 4.97 billion forints (out of a total cost of 6.3 billion forints for ministerial trips abroad) Ministry of Foreign Affairs data, Index, Népszava, 28 September 2026
Criminal complaint by the Ministry of Foreign Affairs on suspicion of breach of trust causing particularly significant financial loss, mid-September 2026 Foreign Minister Anita Orbán (HVG, 28 September 2026)
The presumption of innocence no one may be considered guilty until their guilt has been established by a final court decision Be. Section 1
Limits on informing the public information must be refused, among other things, if it would jeopardise the effectiveness of the proceedings or the unhindered exercise of the rights of those taking part in them Be. Section 109 (1)

One remark belongs with the figures. The private-jet data is not an NVVH case, and this analysis does not claim that it is. It is included here because it illustrates well the significance of a disclosure standard: the same data that was refused a year ago, citing foreign relations, is now public item by item, with its source indicated. The baseline value of an asset recovery case must be just as itemised.

6.3 Policy dimensions

  • Transparency and anti-corruption policy (programme points and background material) — the operating rules of the independent investigation office, the data logic of the public-money dashboard, checks and balances, and the transparency paradox (programme point ID: A1, A6, A10);
  • Justice (programme points) — the reviewability of case-takeover decisions and the guarantees of independence, as well as an impact assessment of the case-card rule at the next amendment of the law on the office (programme point ID: I3, I4);
  • Public administration and e-government (programme points) — results-based efficiency measurement applied to a single office: baseline value, quarterly publication, recovery rate (programme point ID: KI8).

6.4 Literature in detail

6.4.1 Montesquieu: The Spirit of the Laws

Montesquieu assesses confiscation according to the forms of government. In a despotic system, where embezzlement of public money is natural, confiscation is a useful source of revenue for the prince. Under moderate government, however, the situation is quite different:

“Confiscations would render property uncertain, would strip innocent children, would destroy a whole family, instead of punishing a single criminal.”

Montesquieu therefore considers it wise to limit confiscation to specific crimes. Elsewhere he writes that if the prince frequently has the benefit of confiscation, then in deciding criminal causes he would be “both judge and party”. On the reckoning after great political upheavals, he writes that great punishments and great changes necessarily place excessive power in the hands of some citizens. In such cases it is wiser to err on the side of leniency rather than severity, because the task is to eliminate the rebellion, not the rebel. A return to ordinary government, where everyone is protected by the law, should come as soon as possible. Two rules follow from this for Hungarian asset recovery. The first is that recovery should target assets derived from crime, narrowly and on a proven basis, not families and relatives. This is why MIAK asks for the nameless case card in point 3.1. The second is that the recovering office should have no financial interest in confiscation. This is why point 3.3 proposes that the returned amount should not increase the office’s budget.

📖 Source: Montesquieu: The Spirit of the Laws

6.4.2 Robert Klitgaard: Controlling Corruption

As one of the practical lessons of anti-corruption strategy, Klitgaard writes that the public must be convinced that the new effort is serious. According to the Hong Kong and Philippine cases, catching a few top officials and bringing them to trial is an excellent device for this:

“In a corrupt culture pummeling some small fry involved in a small-scale corruption has little effect. Getting serious means, in many settings, frying a big fish in public with accompanying announcements of policy changes.”

Klitgaard also quotes one of the leaders of Hong Kong’s independent anti-corruption commission (ICAC), according to whom the public typically measures the commission’s effectiveness by the rank of the persons prosecuted, which is why the commission needed immediate successes from the outset. This proposition shows both sides of the NVVH’s first four cases at once. On the one hand, it is justified that the first cases carry great weight, because small cases do not break cynicism. On the other hand, if the yardstick of success is the rank of the persons investigated, the office drifts towards spectacular cases, and the measurement itself distorts. According to Klitgaard, catching the big fish also goes together with announcing policy changes. In the NVVH’s case, this change could be that together with today’s announcement the selection and disclosure rules also become public (3.1), and that the yardstick of success is the returned assets, not the rank of the persons investigated (3.3).

📖 Source: Robert Klitgaard: Controlling Corruption

6.4.3 Susan Rose-Ackerman: Corruption and Government

In the part on independent anti-corruption agencies, Rose-Ackerman writes that prosecution services often concentrate on politically less sensitive cases, which is why independent anti-corruption bodies are created. In Hong Kong and Singapore, the turnaround was brought about by the combination of top-level commitment, credible law enforcement by an independent agency based on a strong law, and reform of the civil service. Of the Hong Kong commission (ICAC) she writes that its staff earned more than other civil servants, could not be transferred to other offices, and no one could be placed under a superior who had previously been investigated. According to the author, two measurable signs showed the new institution’s credibility: the growth in the number of complaints it received and the high share of non-anonymous complaints. For the debate on the NVVH’s staff, it follows that staff are protected not by keeping the list of names secret or publishing it, but by institutional rules: protection against transfer, conflict-of-interest rules and traceable decisions. This is what MIAK asks for in point 3.2. And to measure credibility, alongside the recovery rate, the number of complaints received by the office and the share of non-anonymous complaints can also serve as supplementary indicators.

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

6.5 International comparison

The two most frequently cited models, Hong Kong’s ICAC and Singapore’s Corrupt Practices Investigation Bureau (CPIB), both concentrated on large, high-ranking cases in their initial phase. According to the memoir of Lee Kuan Yew (Singapore’s first prime minister), the CPIB was directed at large cases committed at the upper levels, while smaller cases were handled through public, simplified rules and by narrowing discretionary powers. The courts were empowered to confiscate the proceeds of corruption. Behind both models stands a strong office, but in both places credibility came from court results, not from announcements. Two EU legal frameworks also belong here. Directive 2016/343/EU on the presumption of innocence is transposed by the Hungarian Criminal Procedure Act, and this directive also applies to public statements by authorities: they may not refer to a suspect as guilty until their guilt has been established by a final decision. The 2024 EU directive on asset recovery and confiscation also requires Member States to collect statistics on frozen, confiscated and actually recovered assets. The baseline value and recovery rate proposed in point 3.3 are therefore not a Hungarian special path, but an early, case-level application of the EU’s direction on data reporting. Ireland’s office dealing with criminal assets (the Criminal Assets Bureau) reports annually, in a public report, on the amounts secured and the amounts actually paid into the Exchequer. This is the joint publication of the two figures that MIAK proposes in Part IV as a combined yardstick of speed and result.

Transparency and anti-corruption policy

  • A1 — Public-money dashboard
  • A6 — Strengthening checks and balances
  • A10 — Independent Anti-Corruption Investigation Bureau (CPIB model)

Justice

  • I3 — Legislative impact assessment
  • I4 — Protection of judicial independence

Public administration and e-government

  • KI8 — Drucker-style efficiency measurement in public administration

Proposed new programme point: A uniform case card and recovery rate in asset recovery — every public case announcement with six mandatory elements, without names, with a baseline value at announcement, and with accounting for returned assets independent of the office’s budget — for the Transparency and anti-corruption policy area.

6.7 List of sources

Press sources (MIAK topic monitor, 27 September 2026 — topic 1; MIAK press monitor, 28 September 2026 — topic 2):

Knowledge-base references (books):

  • 📖 Montesquieu: The Spirit of the Laws
  • 📖 Robert Klitgaard: Controlling Corruption
  • 📖 Susan Rose-Ackerman: Corruption and Government — Causes, Consequences, and Reform
  • 📖 Lee Kuan Yew: From Third World to First (in the 6.5 international comparison)

MIAK internal materials:

  • MIAK policy area: Transparency and anti-corruption policy (programme points; programme point ID: A1, A6, A10)
  • MIAK policy area: Transparency and anti-corruption policy (background material — the transparency paradox)
  • MIAK policy area: Justice (programme points; programme point ID: I3, I4)
  • MIAK policy area: Public administration and e-government (programme points; programme point ID: KI8)
  • MIAK legal foundations (background material) — delimiting the roles of the prosecution service, the courts and the National Assembly
  • MIAK topic monitor, 27 September 2026 — topic 1, score: 90/100
  • MIAK press monitor, 28 September 2026 — topic 2, score: 84/100

Supplementary public data sources:

  • Act XC of 2017 on Criminal Procedure, Section 1 (the presumption of innocence), Sections 107–109 (informing the public and its limits)
  • Directive (EU) 2016/343 of the European Parliament and of the Council on the strengthening of certain aspects of the presumption of innocence
  • Directive (EU) 2024/1260 of the European Parliament and of the Council on asset recovery and confiscation
  • Criminal Assets Bureau (Ireland) — annual reports

Generation metadata

  • Input press monitor: MIAK topic monitor, 27 September 2026 (supplementary recent sources: MIAK press monitor, 28 September 2026)
  • Generation date: 28 September 2026, 08:20 CEST
  • Tokens used (total): ~105,000 — estimate, blog generation only; the token data of the source topic monitor is not available (see tokens_breakdown in the frontmatter)
  • Translation: Hungarian original at /blog/2026-09-28-nvvh-elso-negy-ugy-kozzeteteli-merce-eljarasi-garanciak-megterulesi-arany/