Part I — Situation overview

On 12 August, after the government meeting held at Paks, Prime Minister Péter Magyar announced that the government has ordered the immediate construction of a Danube bed sill — that is, a stone structure built into the riverbed and remaining below the water surface, which raises the water level of the section above it — between the nuclear power plant and Dunaszentbenedek. With the placement of 35 thousand and then a further 110 thousand cubic metres of stone, the aim is that at the cooling channel the water level should not fall below minus 90 centimetres; the finished structure may raise it by as much as one metre. The first phase takes at least three to four weeks. Until it is ready, a temporary solution is being prepared: two barges, each 80 metres long, will be sunk in a controlled manner, which in the short term may give a rise of as much as 20 centimetres — the first barge is already at Paks, the second on its way, and the preparation of the bank section is under way. The same day the government decree appeared in the Hungarian Official Gazette classifying the riverbed works as a priority investment and the related official cases as of priority importance. The administrative processing deadline is generally 15 days and that of specialist authority procedures 8 days; the environmental operating permit has to be applied for by 30 September. The investment is exempt from obtaining a water permit for construction; a subsequent authorisation for it has to be applied for by 15 October. Neither a water management nor a nature protection fine may be imposed. The tasks were divided between the Bács-Kiskun and the Baranya county government offices. A separate government decision allocated 6.2 billion forints to the National Water Directorate for its damage response tasks against drought and water shortage — this is not the cost of the bed sill, which Portfolio put at roughly 6 billion forints; the decree itself contains no dedicated sum.

The background to the story is that in early August there was a day when generation at Paks hung on a few centimetres. The physical logic of cooling is simple: the four units together produce 6000 megawatts of heat, of which 2000 megawatts becomes electricity, while the remaining 4000 megawatts is carried away by the Danube — the plant draws in 100 cubic metres of water per second and releases it back about ten degrees warmer. The problem is not the flow rate but the water level: the suction pipes of the pumps are fixed at a given level, and if the water surface sinks below this, pumping becomes impossible. The sinking of the riverbed is moreover a decades-long process: according to researchers’ calculations the bed has deepened by about 80 centimetres within a ten-kilometre radius of the plant, primarily because the Austrian and Slovak barrages on the upper Danube section retain the sediment with which the river previously filled its own bed. Without replenishment the river “carves itself out”. The professional debate has meanwhile not ceased: the Paks municipality proposed a barrage and a ship lock near Fajsz, those raising the cooling tower solution invoke heat tolerance, while Magyar Nemzet’s expert compilation highlighted the price and land requirement of cooling towers and their efficiency-reducing capacity loss — 4-5 per cent in warm weather. Attila Aszódi, dean of the Faculty of Natural Sciences of the Budapest University of Technology and Economics, recalled that a hot shutdown of the units is not an extraordinary event, and that a similar water level situation occurred in 2018 as well.

On MIAK’s reading the announcement points in substantively the right direction: the government is handling the cooling water risk of the country’s largest electricity generating unit not with communication but with engineering and budgetary means, and this is the undertaking of intervention rather than its postponement. The package, however, has one striking gap and one striking procedural risk. The gap: the announcement said not a word about the second boundary condition of cooling, the temperature of the Danube, even though under Hungarian regulation the temperature of the river 500 metres downstream of the plant’s warm water discharge point may not reach 30 degrees Celsius — and in June this limit alone took six hundred megawatts off Paks’s output, at a time when there was as yet nothing wrong with the water level. The bed sill raises the water level; it does not carry away the heat; and the impounded section will moreover be deeper and slower, and more slowly flowing water spends more time in the sun and mixes worse, so there is a theoretical possibility that dilution deteriorates precisely where the official measuring cross-section is. The risk: the priority investment legal instrument is precisely that accelerated procedural form whose abusive use MIAK consistently criticised in the previous term — and which the new government has now used for the first time. MIAK’s credibility depends on applying the same yardstick: urgency may justify a shortened procedure, but it does not extinguish the obligation of ex-post impact assessment and publicity.

Part II — Literature foundation

Three sources provide the interpretative frame. In the adaptation chapter of the Stern Review, the British economist Nicholas Stern, head of the British Treasury’s climate economics report, introduces the concept of maladaptation: a measure taken in response to a climate impact which in fact increases vulnerability — and whose most frequent cause is that the decision-maker does not know or does not take into account the effects of their own measure spilling over into other sectors. In the Paks case this concept is directly applicable: a water-level-raising structure whose temperature and flood consequences have not been modelled may slip into exactly this category. In her work Governing the Commons, the American political economy researcher Elinor Ostrom, awarded the Nobel Memorial Prize in 2009 for her research on the management of common resources, works up the decades-long cases of the Californian groundwater basins and records two patterns that hold here too: stock depletion processes are self-reinforcing and at a certain point irreversible, and successful interventions almost without exception began with an independent technical survey, not with the intervention. In analysing decision-making, the Austrian-American management thinker Peter F. Drucker, founder of the methodology of organisational efficiency measurement, names the clarification of boundary conditions as the precondition of an effective decision: it has to be stated in advance what must be fulfilled, failing which the decision cannot be regarded as successful even if it is carried out. The Paks announcement was silent about precisely one boundary condition — the temperature limit. The detailed treatment of the literature — author by author, with quotations — can be found in the 6.4 Literature in detail section.

📖 Source: Nicholas Stern: The Economics of Climate Change — The Stern Review; Elinor Ostrom: Governing the Commons — The Evolution of Institutions for Collective Action; Peter F. Drucker: The Effective Executive

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. None of them aims at stopping the intervention — the intervention is justified, and delaying it is in itself a risk. All three proposals aim to ensure that the accelerated procedure does not come with the loss of measurement and accounting.

3.1 A public hydrodynamic and heat propagation model before the start of the second phase (30 days)

Permitting currently runs in parallel with construction, which in a crisis situation is a defensible decision — but the parameters of the structure are normally determined by a detailed hydrodynamic model: the size of the stone covering against downstream scouring, the geometry of the sill and the extent of flood backwater. MIAK proposes that before the start of the second, 110 thousand cubic metre phase — at the latest within thirty days — a model calculation be published that gives an account of three questions: how much backwater the structure causes at flood discharge, expressed in centimetres; by how much flow velocity falls on the impounded section, and what temperature rise this means at the official measuring cross-section under summer mean and low water conditions; and finally how much downstream bed deepening is expected below the sill on a five- and ten-year horizon. The claim that “during a flood this will pose no danger whatsoever” is not self-evident — a submerged sill always causes some backwater; it may be negligible, but this has to be calculated. If the calculation is already available from earlier expert studies, then the proposal amounts to publishing it. The principle of legislative impact assessment (I3) prescribes precisely this: a public statement of what a measure solves, whom it affects and what it costs.

3.2 Including the temperature limit among the boundary conditions of the decision

MIAK proposes that the government supplement the Paks cooling measure package with a timetable of the same level of detail addressing the temperature limit, and that it state this boundary condition publicly. Its content has three elements. First: there should be public, continuously updated data publication on the Danube temperature measured at the cross-section and on how many megawatts of capacity loss the heat limit caused in the given month — this is not available today in a form the public could follow. Second: a cost comparison should be produced of the possibilities for lastingly handling the heat problem — deeper water intake, a cooling tower, or maintaining the present state — expressing the investment cost and the efficiency loss together, in forgone generation and in forints. The difference between the cooling tower and deeper water intake is decisive here: deeper water intake solves only the water level problem, the cooling tower solves both, because it transfers the residual heat to the air — but it is more expensive, needs more space and reduces efficiency in warm weather. MIAK does not take a position on the technical solution; it asks that the decision follow from a comparison supported by figures. Third: the development of climate modelling capacity (K9) should include the joint forecasting of Danube water temperature and discharge, because this is the input from which the planning of the energy system’s shock resilience (K7) can be produced at all.

3.3 The price of the accelerated procedure: an ex-post impact assessment with a fixed deadline and public cost accounting (12 months)

MIAK proposes that the government voluntarily attach to the priority investment decree — even in a separate decision — a public ex-post impact assessment to be completed within twelve months of the completion of the investment. This is the yardstick MIAK formulated against the accelerated procedures of the previous term, and which it now applies in the same way after a change of government. The impact assessment should answer four questions: whether the technical goal was met, that is, whether the water gauge remained above the minus 90 centimetre level; how much the investment actually cost, in an itemised breakdown, and how this relates to the roughly 6 billion forints mentioned at the time of the announcement; what actual environmental effect is measurable compared with the forecast of the review containing the Natura 2000 impact assessment; and how much forgone generation, or replacement cost, the intervention saved. This structure is the compulsory application of the Drucker audit — that is, the ex-post comparison of expected versus actual effect of the measures taken — to significant state measures (G20), while the public publication of the cost items follows from the principle of the public money dashboard (A1). A separate request is that the accounting of the 6.2 billion forint drought damage response fund also feature in the same place: this sum — the first case in which the government made an exception to the finance minister’s veto right fixed in May — arrived on a different legal title and for a different purpose, and conflating the two would afterwards make the evaluation of both impossible.

The three proposals are held together by a single principle: a crisis situation justifies the speed of the decision, not the suspension of its accountability. Stern’s concept of maladaptation (see 6.4.1) warns that bad adaptation is a question not of intent but of missing information — so the model calculation is not a bureaucratic extra step but the only way of avoiding the error. Ostrom’s Californian cases (see 6.4.2) show that the independent technical survey paid off even when the actors were under time pressure. And Drucker’s concept of boundary conditions (see 6.4.3) tells us what the missing temperature limit means: not a matter of detail, but the condition without whose fulfilment the decision is not successful even if the stone goes into place.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Energy supply The water level can be kept above the pumping threshold, so a full shutdown and the purchase of more expensive imported electricity can be avoided The temperature limit remains binding as before; there is a theoretical possibility that because of the slower flow of the impounded section dilution deteriorates precisely at the measuring cross-section, and that next summer the plant cannot run at full capacity even with a satisfactory water level
Environment Alongside the handling of the immediate risk, carrying out the review containing the Natura 2000 impact assessment remains compulsory Below the sill bed deepening typically accelerates, while above it sediment is deposited — a trap well known in hydraulic engineering is that one sill brings with it the need for the next; and the exclusion of fines moreover removes the instrument of discipline during implementation
Public administration The accelerated procedure does genuinely shorten lead time, and the task is clearly divided between two government offices The priority investment form now sets a precedent in the new government’s practice: if it is not accompanied by a voluntarily undertaken ex-post impact assessment, in the next, less urgent case this precedent will already be the reference point
Shipping and infrastructure After completion of the second phase, the raised water level may also improve the conditions for restarting shipping Continuous monitoring, maintenance and possible subsequent construction are a lasting cost item that does not appear in the announced sum

The most important question to weigh lies between urgency and measurement, and it would be dishonest to pretend that it is not real. If the intervention waited for the full permitting and modelling cycle, in the autumn falling-water phase the plant could again come close to shutdown — the cost of delay is therefore not theoretical. If, on the other hand, no public model and ex-post accounting accompanies the acceleration, then two separate harms arise: the technical one, if the sill is badly dimensioned or backfires on the heat side, and the institutional one, because the use of the priority investment form becomes a precedent. The proposal tips to the risk side if the publicity obligation is interpreted as requiring the works to be suspended until the model is completed — that is not the request. The proposal ties the publication of the model to the start of the second phase, precisely because the first, 35 thousand cubic metre phase is justified by response time, whereas the second is already plannable.

Part V — Measurability and summary

5.1 What is worth tracking? (proposed KPIs)

MIAK proposes four performance indicators (KPIs) for tracking:

  • Whether the Paks water gauge stays above the minus 90 centimetre level — daily publication is proposed until the start of the heating season. This is the stated technical goal of the intervention, and therefore the primary yardstick.
  • Capacity loss due to the heat limit, monthly, in megawatts — monthly publication is proposed for the summer months. This is the indicator that shows how large a residual item the heat problem is after the water level problem has been solved; in June it was six hundred megawatts.
  • The actual investment cost compared with the announced order of magnitude — itemised publication after completion is proposed. It is worth tracking separately the accounting of the bed sill and of the 6.2 billion forint drought damage response fund, because the two arrived on different legal titles.
  • The development of the bed level on the section below the sill — annual measurement is proposed. This reveals whether the accelerating downstream deepening that would bring with it the need for the next sill has begun.

5.2 Summary

MIAK’s request can be summed up in three points, all three with deadlines: before the start of the second, 110 thousand cubic metre phase, the model calculation on flood backwater, flow velocity and downstream scouring should be published; the temperature limit should be included among the stated boundary conditions of the package, with a timetable of the same level of detail and public data publication; and alongside the priority investment form the government should voluntarily attach a public ex-post impact assessment within twelve months of completion. From the public MIAK asks that the question not be read in an “intervention versus environmental protection” frame: the intervention is justified, and the real subject of the debate is whether measurement survives alongside urgency.

In this matter two of MIAK’s foundational values move. Accountability, because the accelerated procedural form always offers the same trade: it gains time and gives up a procedural guarantee. This trade is defensible in a crisis — but only if the lost prior guarantee is replaced by ex-post accounting with a fixed deadline. If not, then acceleration is not a trade but a loss. And data-drivenness, because in this matter there is a measure for only one of the decision’s two boundary conditions — the water level — and none for the other, the temperature of the water; and as long as there is no regular, public data on the loss caused by the heat limit, the public cannot speak meaningfully about its extent. The yardstick MIAK applied to the previous government’s accelerated investments is the same now — this is the only test of consistency.

Part VI — Justifications and further sources

6.1 The press framing by spectrum

The left-liberal and public affairs band this time chose not a political but an explanatory frame, which is rare and fortunate from the point of view of the topic. Of Telex’s two pieces, one unpacked the procedural content of the Official Gazette decree item by item (deadlines, permit exemptions, exclusion of fines), while the other was expressly a concept-explaining article: “What is a bed sill, and how does it save the nuclear power plant?” — with an expert who described the structure with the analogy of stones placed in a mountain stream, and who also stated that the extent of the flood water level rise depends on the parameters, which are not yet known. HVG reported the facts of the decree and the announcement, and it is the only source that separately highlighted the 6.2 billion forint water management damage response fund and the fact that with this the government made an exception to the finance minister’s veto right for the first time. 444.hu arranged the technical parameters and the forecast figures in chronological order, and it alone drew the regional parallel: a few days earlier, in Romania, four barges loaded with stone were sunk in one branch of the Danube to secure the water supply of the Cernavodă nuclear power plant.

The economic band gave the deepest analysis, and at the same time the only substantive criticism. Portfolio’s piece “What is the Paks rescue package enough for — and what not?” separated the three boundary conditions of cooling (water level, discharge, temperature), and concluded that the announced package answers one of them; this article brought the June six hundred megawatt item of the heat limit, the explanation of the decades-long, eighty-centimetre sinking of the bed level, and the possible side effect according to which on the impounded, more slowly flowing section dilution may deteriorate precisely at the official measuring cross-section.

The pro-government conservative band did two different things on this day. Magyar Nemzet brought an expert compilation on the technical and cost-side limits of the cooling tower alternative, and reported on the Paks municipality’s barrage proposal near Fajsz — that is, it stayed in a substantive, not a political frame. Mandiner, by contrast, put a political dispute on its front page: the response of a Fidesz politician to the prime minister having, in their view, questioned their professional credibility, and who also claims that the temporary water diverter now ordered was their own earlier proposal. ATV gave a third voice: from Attila Aszódi’s remarks it highlighted that a hot shutdown of the units is not an extraordinary event, and that in his view the summer’s experience also supports the necessity of the Paks II project. Looking at the spectrum as a whole, what is striking is that the question of the temperature limit appeared in a single band, the economic specialist press — the political debate was about the water level and the authorship of the intervention.

6.2 Facts and data

Datum Value Source
Stone to be placed for the bed sill 35,000 in the first round, then a further 110,000 cubic metres (approx. 145,000 m³ in total) Péter Magyar’s announcement; Portfolio, 444.hu, HVG, 12–13 August 2026
Order-of-magnitude cost of the investment approx. 6 billion forints Portfolio, 13 August 2026
Lead time of the first phase at least 3-4 weeks Péter Magyar’s announcement; HVG
Target water level at the cooling channel should not fall below minus 90 cm; the finished structure may raise it by as much as 1 metre Péter Magyar’s announcement; 444.hu, HVG
The temporary solution controlled sinking of two 80-metre barges, a rise of as much as 20 cm Péter Magyar’s announcement; HVG, 444.hu
Forecast water level without intervention approx. minus 136–137 cm within 4-5 days, and below minus 140 cm the following week water management forecast as presented by Péter Magyar; 444.hu
Thermal and electrical output of the four units 6000 MW of heat, of which 2000 MW electricity; 4000 MW residual heat into the Danube Portfolio, 12 August 2026
The plant’s cooling water abstraction approx. 100 m³/s, released back approx. 10 °C warmer Portfolio
The temperature limit 500 m downstream of the discharge point the Danube may not reach 30 °C Hungarian regulation; Portfolio
Loss due to the heat limit in June approx. 600 MW Portfolio
Sinking of the bed level within a 10 km radius of the plant approx. 80 cm over recent decades researchers’ calculations; Portfolio
Drought and water shortage damage response fund for the National Water Directorate 6.2 billion forints (separate government decision; the first exception to the finance minister’s veto right) Hungarian Official Gazette; HVG, 12 August 2026
Priority investment procedural deadlines administrative processing generally 15 days, specialist authority procedure 8 days Hungarian Official Gazette; Telex
Permitting deadlines application for the environmental operating permit by 30 September 2026; water subsequent authorisation and nature protection permits by 15 October 2026 Hungarian Official Gazette; Telex, HVG

Two remarks for interpretation. First, sources gave the unit of the cubic metre figure differently: some reports gave the 35 thousand and 110 thousand items in tonnes, others in cubic metres. In the table MIAK took as a basis the cubic metre figures of Portfolio and 444.hu, because these two sources also give the full 145 thousand cubic metre total — but the unit of the item cannot be publicly verified from the decree, and this in itself indicates the absence of public technical documentation. Second, 6 billion and 6.2 billion forints are two separate items: the first is the order-of-magnitude cost of the bed sill investment on the basis of press information, the second the sum set aside for water management damage response and fixed in a government decision. Because of the similarity of the two figures, conflation in everyday speech is obvious, and precisely for this reason proposal 3.3 asks for both to be accounted for separately.

6.3 Policy dimensions

  • Environment and climate (programme points and background material) — energy market shock resilience (programme point ID: K7) provides the frame in which a cooling water constraint is not a weather event but an energy security item; the domestic integrated climate economics model (programme point ID: K9) provides the modelling capacity from which the joint forecasting of Danube discharge and temperature can be produced at all; while the principle of measurable climate targets (programme point ID: K1) prescribes that targets be numerical and publicly trackable — the minus 90 centimetre water level target is precisely such a target, and therefore has to be made trackable;
  • Economy (programme points) — the economic policy impact assessment system (programme point ID: G20) provides the direct basis of proposal 3.3: compulsory ex-post impact assessment for every significant measure, after 12-18 months;
  • Justice (programme points) — the principle of legislative impact assessment (programme point ID: I3) is what the demand for a public model calculation follows from: legislation — here the priority investment decree — has to show publicly what it solves, whom it affects and what it costs;
  • Transparency and anti-corruption policy (programme points) — the principle of the public money dashboard (programme point ID: A1) provides the requirement of itemised, public publication of cost items, which is especially important in the case of an accelerated procurement procedure;
  • Public administration and e-government (background material) — it provides the interpretative frame of the procedural regime for cases of priority importance and of the division of competences between the two government offices.

6.4 Literature in detail

6.4.1 Nicholas Stern: The Economics of Climate Change — The Stern Review

The chapter of the Stern Review dealing with adaptation starts from the fact that in the coming decades adaptation is the only instrument for moderating the already unavoidable climate impacts — but adaptation is itself costly, and does not cancel out the impacts. In this frame it introduces the concept that is the name of the central risk of the present decision:

“Some measures will not be favourable from a climate change perspective and in some cases will increase vulnerability to the impacts of climate change. This is termed maladaptation. Maladaptation is often caused by a lack of information about, or a failure to take account of, the external effects that regulations and practices have on other sectors.”

At the same point the report also records that separating out adaptation costs is methodologically difficult, because adaptation is so wide-ranging and cross-cutting — it touches economic, social and environmental conditions at once.

To the Paks case this concept fits almost word for word. The bed sill is the adaptive answer to the water level, but it also affects two other “sectors”: the river’s temperature regime, through the slower flow of the impounded section, and flood conveyance, through backwater. On Stern’s definition maladaptation typically derives not from intent but from missing information — that is, the question is not whether the decision-maker acts in good faith, but whether the calculation describing the effects is available to them. This provides the theoretical justification of proposal 3.1: the model calculation is not an argument against the intervention but precisely what protects the intervention from causing harm. And it also provides the justification of proposal 3.2: if the temperature limit is left out of the decision’s calculation, then in Stern’s category we have exactly the case of “a failure to take account of”.

📖 Source: Nicholas Stern: The Economics of Climate Change — The Stern Review

6.4.2 Elinor Ostrom: Governing the Commons — The Evolution of Institutions for Collective Action

The middle chapters of Ostrom’s volume work up the decades-long cases of the Californian groundwater basins — situations in which the depletion of a jointly used water stock threatened physical irreversibility, and in which the actors had to find an institutional answer under pressing time constraints. Two of her observations apply directly to the Paks situation. The first is the self-reinforcing nature of the depletion process:

“…the gravel and sand in the aquifers eventually become so compacted that they can no longer hold as much water as before.”

The other is the order of intervention. In Ostrom’s description, in the successful Californian cases the actors had an independent survey carried out before the legal and technical step: the water users’ association of the Central Basin “employed a private engineering firm with well-known expertise in the field of groundwater basins to carry out the first survey of the condition of the basin and of past water use”, and they made a considerable effort to reach consensus in advance on the type of agreement.

The Hungarian parallel is close at two points. On the one hand, the sinking of the Danube bed is just as self-reinforcing and physically irreversible a process as the one Ostrom describes: the barrages of the upper section retain the sediment, without replenishment the river deepens its own bed, and hydraulic engineering experience shows that below a bed sill deepening typically even accelerates — that is, “one sill brings with it the need for the next”. The present intervention is therefore not a conclusion but the beginning of a lasting maintenance and monitoring obligation, and this has to be shown as such on the cost side too. On the other hand, the main lesson of Ostrom’s cases is procedural: urgency did not extinguish the need for an independent survey there either, but rather provided the basis for rapid agreement. This is the practical reinforcement of proposal 3.1 — the model calculation does not slow the intervention but underpins it.

📖 Source: Elinor Ostrom: Governing the Commons — The Evolution of Institutions for Collective Action

6.4.3 Peter F. Drucker: The Effective Executive

In his chapter on effective decision-making Drucker derives the quality of a decision not from the result but from the degree of clarification before the decision. The central concept is the boundary condition: the statement of what has to be fulfilled for the decision to count as a solution at all.

“The more concisely and clearly boundary conditions are stated, the greater the likelihood that the decision will indeed be an effective one and will accomplish what it set out to do. Conversely, any serious shortfall in defining these boundary conditions…”

Drucker’s examples are about why stating boundary conditions is critical: they can decide the structure of the solution — with a badly or tacitly formulated boundary condition, implementation may be flawless and the decision still miss its target.

The Paks announcement is deficient at precisely this point. The boundary condition stated was one: the water level should not fall below minus 90 centimetres. This is clear, measurable and trackable — in this the announcement is good. Cooling, however, according to public professional analyses has three boundary conditions: water level, discharge and water temperature. If the decision states only the first, then in Drucker’s conceptual frame the decision may “miss its target” even if the stone goes fully into place — because in a 34-degree heatwave, with a 28-degree Danube, the water level may be fine and the plant still unable to run at full capacity. This would be a fault not of implementation but of decision preparation. MIAK’s proposal 3.2 therefore does not ask for a new technical solution but for the statement of the missing boundary condition and the timetable belonging to it — in Drucker’s formulation this is the step that makes a decision effective. The same author also provides the theoretical basis of proposal 3.3: the Drucker audit in programme point G20 institutionalises precisely the back-measurement that sets expected against actual result.

📖 Source: Peter F. Drucker: The Effective Executive

6.5 International comparison

The curtailment of river-cooled nuclear power plants because of low water levels and warming is not a Hungarian peculiarity, and the answers can be arranged into three models. The French model has rested on regulatory flexibility: in the heatwaves of recent decades the French authority has on several occasions granted temporary exemptions from river water temperature discharge limits in order to preserve security of supply. The advantage of this model is speed; its disadvantage is that the ecological limit value becomes the adaptation variable, which in the long run is at the expense of the river’s wildlife, and which moreover does not solve the problem but only postpones it. The closed-cooling model — Mochovce in Slovakia and Dukovany in the Czech Republic — transfers the heat to the air with cooling towers by evaporation: these too need make-up water, but the fresh water requirement is a fraction of that of open systems, so they are substantially more resistant to heat. The price consists of the investment cost, the larger land requirement, more complex operation, higher maintenance needs and a 4-5 per cent efficiency loss in warm weather. The riverbed intervention model is operating right next door: in Romania, a few days before the Hungarian decision, four barges loaded with stone were sunk in one branch of the Danube to secure the water supply of the Cernavodă plant, in order to divert more water towards it — the present Hungarian barge solution is a close counterpart of this.

Of these three models the Hungarian decision chose the third for immediate handling, and the lasting variant of the third for the medium term. This is a defensible choice, because it is the fastest and requires the least investment. Two lessons, however, follow from the international picture. One is that the riverbed intervention model has not solved the question of the temperature limit in any country — to that the answer is either regulatory exemption or closed cooling, and between the two a political rather than a technical decision is needed. The other is that the closed-cooling systems operate precisely in the region whose hydrological exposure most resembles the Hungarian — that is, the cost comparison of the cooling tower alternative is not a theoretical question but work that can be done, and for which the data of neighbouring operators are also available. This is the practical basis of the second element of proposal 3.2.

Environment and climate

  • K1 — Measurable climate targets
  • K7 — Energy market shock resilience
  • K9 — Domestic integrated climate economics model (DICE-HUNGARY)

Economy

  • G20 — Economic policy impact assessment system (Drucker audit)

Justice

  • I3 — Legislative impact assessment

Transparency and anti-corruption policy

  • A1 — Public money dashboard

Proposed new programme point: An ex-post impact assessment obligation with a deadline tied to accelerated investment procedures — for the Public administration and e-government area.

6.7 List of sources

Press sources (MIAK press monitor, 13 August 2026 — topic 1):

Knowledge-base references (books and official analyses):

  • 📖 Nicholas Stern: The Economics of Climate Change — The Stern Review
  • 📖 Elinor Ostrom: Governing the Commons — The Evolution of Institutions for Collective Action
  • 📖 Peter F. Drucker: The Effective Executive

Note: the local file path of the sources does not appear in the visible text of the blog — only the author and the title.

MIAK internal materials:

  • MIAK policy area: Environment and climate (background material and programme points; programme point ID: K1, K7, K9)
  • MIAK policy area: Economy (programme points; programme point ID: G20)
  • MIAK policy area: Justice (programme points; programme point ID: I3)
  • MIAK policy area: Transparency and anti-corruption policy (programme points; programme point ID: A1)
  • MIAK policy area: Public administration and e-government (background material)
  • MIAK press monitor, 13 August 2026 — topic 1, score: 94/100

Additional public data sources:

  • National Water Directorate — Danube water level and discharge time series, Paks water gauge
  • MAVIR — system load and generation data
  • Hungarian Energy and Public Utility Regulatory Authority — capacity and security of supply reports
  • Joint Research Centre of the European Commission — European Drought Observatory
  • Hungarian Official Gazette — the text of the priority investment government decree and of the water management damage response government decision

Generation metadata