Part I — Situation overview

On 13 September 2026 two senior research fellows of the Regional Centre for Energy Policy Research (REKK, a research unit specialising in energy-market analysis) published in the opinion section of G7 what the summer curtailment of the Paks nuclear power plant had cost. The calculation covers the period from the brief 160-megawatt load reduction on 18 July, caused by the Danube’s water temperature, to the restriction tightened in several steps that lasted until 26 August. Its starting point is that the entire output of the Paks Nuclear Power Plant is sold by the MVM group’s trading company, so MVM had to buy the undelivered electricity on the market in order to meet its own contractual obligations. The authors multiplied the lost volume by the quarter-hourly spot wholesale prices and subtracted the plant’s production cost of 14 forints per kilowatt-hour, derived from its 2025 annual report. The result: on some days the loss exceeded two billion forints, and for the period as a whole it came to 37.4 billion forints. The authors regard this as an upper estimate, because during the prolonged outage the trader may also have combined its purchases with cheaper short-term products. The occasion for the calculation is the government resolution published on 3 September, which is an internal governance decision of the government, not a piece of legislation. With a deadline of 11 September, the resolution ordered a comprehensive report on the circumstances that led to the operating situation at Paks, with the involvement of the Hungarian Atomic Energy Authority (OAH) and MVM.

So far there is no trace of the report being made public in the downloaded sources. MIAK has already followed the background of the Paks events in several analyses — in connection with the full shutdown and the lack of a reconnection protocol, and then the restart of Unit 3 and the hydrological planning yardstick, where MIAK asked for an itemised final report on the crisis management. Now a new question has opened up: the bill. The same weekend the second wave of the energy-price shock also began. On 11 September, after drone attacks, Saudi Arabia closed as a precaution the 1,200-kilometre East–West crude oil pipeline that bypasses the Strait of Hormuz and in recent months carried 4–5 million barrels a day, 4–5 per cent of global supply. According to Reuters sources, Yemen’s Houthis seized Perim Island in the Bab el-Mandeb strait at the entrance to the Red Sea. The price of Brent crude jumped to close to 110 dollars a barrel on Thursday, before settling at around 104–105 dollars.

MIAK’s reading: the two news items point to the same gap. The Paks outage and the pipeline shutdown are both risks that could have been foreseen — the low-water risk was also recorded in the OAH’s 2025 national report, and the vulnerability of the Red Sea route has been documented since 2024. Unpreparedness therefore has not only a technical but also a financial price, and that price is now looking for someone to pay it in the absence of a rule. As long as it is not fixed in advance who bears the damage arising from a foreseeable risk, the loss quietly migrates from the owner to the budget or to the consumer — and nobody is incentivised to prepare for the next shock either.

Part II — Foundations in the literature

Three sources provide the frame. Adam Smith (Scottish philosopher and economist, founder of modern economics) observes in The Wealth of Nations (1776), writing about joint-stock companies, that those who manage other people’s money do not watch over it with the same vigilance as over their own — the lack of risk management is thus not a flaw of character but an incentive structure, which only a rule fixed from outside can counterbalance. The International Monetary Fund’s (IMF) World Economic Outlook 2024 documented two years ago that the Red Sea attacks cut traffic through the Suez Canal by almost two-thirds. The report also stated that the world economy is now dominated by supply disruptions, whether of climate, health or geopolitical origin — that is, the low water at Paks and the shutdown of the Saudi pipeline belong to the same risk class. Ray Dalio (American investor, founder of Bridgewater Associates) makes the open, objective examination of mistakes and the distinction between proximate causes and root causes a condition of organisational learning in Principles — this is the argument for publishing the report ordered by the government resolution. The detailed treatment of the literature — by author, with quotations — is in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. None of them is about naming those responsible after the fact; they are about ensuring that at the next outage it is known in advance who pays and why.

3.1 Public release of the report under the government resolution, with a cost chapter (within 30 days)

MIAK asks that the report completed by the 11 September deadline also be published within 30 days — with the redactions justified on national-security and nuclear-safety grounds. The final report MIAK requested in its analysis of 23 August 2026 would have covered the technical and investment side of the crisis management; the present request supplements it with a separate cost chapter containing four items. The first is MVM’s actual cost of the lost output and the replacement purchases, compared with REKK’s upper estimate. The second is a statement of the channel through which this cost appears: in MVM’s own profit, in a reduced dividend paid to the owner, in a possible capital injection, or in universal-service prices and their supplementary funding. The third is an assessment of whether the risk was foreseeable: whether the low-water risk appeared in official documents before the report, and if so, which risk-reduction measure was not taken. The fourth is the separation of root causes from proximate causes (see 6.4.3). This is a direct application of the G19 radical transparency programme point and the G1 data-driven budget: a loss affecting public money, whose order of magnitude is known, cannot remain an internal document.

3.2 An outage cost-bearing rule fixed in advance for state-owned energy companies (by the first quarter of 2027)

The second proposal concerns the future. MIAK proposes that, on a proposal by the Government — at the level of an act or a government decree, fitted into the framework for exercising ownership rights — a rule be adopted by the first quarter of 2027 on who bears the additional cost arising from the outage of strategic generating units. The rule should rest on a single, verifiable criterion: whether the risk appeared, before the event, in a public official document (an authority report, a security-of-supply assessment, the company’s own risk register). If it did, the damage is borne by the company and its owner — it cannot be passed on into universal-service prices or into the budgetary funding that supplements them. If the risk was not known in advance, the cost may be socialised, but only on a separate, named budget line, not hidden in prices. The Hungarian Energy and Public Utility Regulatory Authority (MEKH), as the regulator, should check during price preparation that the excluded cost does not appear indirectly. According to Smith’s thesis (see 6.4.1) the vigilance of management handling other people’s money is ensured not by good intentions but by consequences — under state ownership, where the owner is the community itself, the rule has to supply this consequence.

3.3 A unified shock-cost statement and the severe scenario as the planning baseline (by the submission of the 2027 budget)

The third proposal links the two shocks. MIAK proposes that the G25 energy-price shock preparedness plan be completed by the submission of the 2027 budget, and that it treat the severe version of a loss of the Red Sea route not as an extreme case but as the planning baseline (see 6.4.2). The plan should be accompanied by an annually updated public shock-cost statement showing in a single table what the energy-market shocks of a given year cost and who paid for them: the state-owned company, the budget or the consumer. This table would include the 37.4 billion forints of Paks and the roughly 6 billion forint budgetary burden of the diesel support. The eligibility and phase-out questions of the diesel support were discussed in MIAK’s analysis of 8 September 2026; this piece highlights only one new element. For the road-haulage sector, which accounts for more than half of domestic diesel consumption, a pre-priced liquidity instrument could also be part of the plan instead of price support — for example a time-limited, repayable deferral of excise duty. Its budgetary impact is an interest burden and a default risk, not non-repayable expenditure. The automatic mechanism of the K7 energy-market shock resilience programme point provides the thresholds for this.

The three proposals are bound together by one principle: the price of a shock falls if it is known in advance who pays it. If the owner knows that the damage arising from a foreseeable risk stays with it, then redesigning the cooling system or relocating the pumps — which according to the REKK authors would have cost far less than barely a month of losses — becomes a profitable investment, not a cost that can be postponed.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Budget The shock-cost statement makes hidden burdens visible; the named line rules out the loss dissolving into price supplements Under state ownership the damage remaining with the owner is ultimately also public money (lower dividend, capital injection) — the rule does not eliminate the burden, it only makes it transparent and incentivising
Consumers The damage from foreseeable risks is not built into universal-service prices The exclusion can be circumvented by indirect pass-through (for example by raising other cost lines) if MEKH’s checks remain a formality
State-owned enterprise governance Risk-management investments become profitable because the consequence of neglect is known in advance The criterion “did the risk appear in a document” can create a perverse incentive: the company may record fewer risks so that it can later invoke unforeseeability
Energy security Making the severe oil-market scenario the planning baseline prepares targeted instruments in good time Excessive stockpiling and preparation are costly if the shock fades quickly

The most sensitive point is the perverse incentive in the third row. If cost-bearing depends on whether the risk appeared in a document, the company may have an interest in documenting less. Two elements counterbalance this: the criterion extends not only to the company’s own register but also to authority and security-of-supply reports — in the case of Paks it was precisely the OAH report that recorded the risk — and deficiencies in the risk register are in themselves a matter of the owner’s responsibility. The second question for deliberation is timing: the rule cannot be applied retroactively to the summer loss, so the fate of the 37.4 billion forints requires an individual but publicly justified decision on the basis of the report under point 3.1. The proposal tips towards the risk side if, instead of publicity, internal accounting prevails again and at the next outage the “extraordinary event” argument decides once more.

Part V — Measurability and summary

5.1 What is worth following? (suggested KPIs)

MIAK proposes four performance indicators (KPIs, in English: Key Performance Indicator) from which it will be visible in 6, 12 and 24 months whether the proposal has hit its target:

  • Publicity of the report under the government resolution: the suggested target is that the report, together with its cost chapter, be publicly available by mid-October 2026.
  • Existence of the cost-bearing rule: the suggested target is that by the first quarter of 2027 a promulgated rule fix how the cost of foreseeable and unforeseeable risks is borne.
  • Statement of the cost of the Paks outage: it is worth following whether MVM’s 2026 annual report shows the additional cost caused by the outage separately and as an amount, and whether it can be compared with the upper estimate of 37.4 billion forints.
  • The shock-cost statement and the severe scenario: the suggested target is that the preparedness plan under G25 appear as an annex to the 2027 budget proposal, including the scenario of losing the Red Sea route and the cost estimate of the liquidity instrument for hauliers.

5.2 Summary

MIAK’s request in brief: the Paks bill should not quietly find someone to pay it. Concretely: the report required by the government resolution should appear within 30 days, with a cost chapter; by the first quarter of 2027 a rule should be adopted under which the damage from a foreseeable risk is borne by the owner, and that from an unforeseeable one by a named budget line; and for the 2027 budget a preparedness plan based on the severe oil-market scenario should be completed, with an annually updated shock-cost statement. The second wave of the oil supply shock shows that the next outage is only a matter of time.

Two MIAK foundational values are at stake. One is accountability: the upper-estimate loss of 37.4 billion forints at a state-owned company is more than one and a half times the plant’s entire 2025 operating profit — no decision on how to handle such an item can be taken without public justification. The other is data-drivenness: the cost-bearing rule rests not on the political assignment of responsibility but on a verifiable fact — whether the risk appeared in official documents in advance. In this way the yardstick remains the same for every future government and every future outage.

Part VI — Reasoning and further sources

6.1 Press framing by spectrum

In the monitored Hungarian press the question of the additional cost of Paks was raised only by G7, part of Telex, and in its opinion section, written by outside authors. The framing is explicitly normative: the article does not merely calculate the loss but also takes a position that the low-water situation cannot be regarded as an extraordinary, unforeseeable event, and therefore society should not bear the damage. None of the news portals picked up or disputed the calculation; so far the topic rests on a single expert text in public debate.

The economic press discussed the oil-market shock in a global market frame. In three articles Portfolio presented the military background of the pipeline closure, the central-bank reactions — the European Central Bank has raised rates for the second time since the outbreak of war in February — and a corporate counter-example: Poland’s Orlen announced that thanks to diversified sourcing the shutdown does not affect the supply of its refineries. The same day the paper also wrote about the dynamic household electricity tariff and, quoting an expert, raised a more transparent separation of subsidy and market price — an idea related to the consumer side of the Paks cost-bearing question. 24.hu put the emphasis on the geopolitical stakes: according to Máté Szalai’s analysis, 10–12 per cent of global seaborne trade and the undersea cables carrying around 90 per cent of Europe–Asia internet traffic pass through Bab el-Mandeb, and Europe “sleepwalked” into this situation. An earlier 24.hu article highlighted that the Paks riverbed sill does not solve the evening import dependence of domestic supply.

HVG linked two threads: the diesel-price impact of the Middle East escalation and the outrage of hauliers’ associations. With a balancing remark it also noted that the support is cheap and less harmful to retail than the earlier price regulation, only it leaves out large consumers. The government-critical Magyar Nemzet framed the severity of the shock (“Saudi Arabia’s last reserve oil pipeline”) and the lack of credibility of the government’s calculation — citing surveys on refuelling frequency it disputed the assumption of refuelling once a month, and highlighted the hauliers’ “united front”. ATV gave a factual, service-oriented account of the finance minister’s briefing. 444.hu and Népszava did not carry the topic among the monitored headlines that day; the headlines of Index and Mandiner could not be evaluated.

6.2 Facts and data

Datum Value Source
The government resolution ordering the Paks report published 3 September 2026; reporting deadline 11 September, with the involvement of OAH and MVM REKK authors, G7, 13 September 2026
The curtailment period examined 18 July 2026 (160 MW load reduction) – 26 August REKK authors, G7
Threshold for hours counted as outage / normal output level hours below 1,750 MW / 1,850 MW REKK authors, G7
Paks plant’s 2025 revenue / operating profit / output HUF 247.9 billion / HUF 22.2 billion / 16,097.9 GWh Paks Nuclear Power Plant 2025 annual report, as reported by the REKK authors
Calculated production cost HUF 14.02/kWh (HUF 14/kWh in the calculation) REKK authors, G7
Additional cost for the whole period HUF 37.4 billion (upper estimate); above HUF 2 billion on some days REKK authors, G7
Ratio of the additional cost to 2025 operating profit around 1.7 times MIAK’s own calculation from the above data
Throughput of the East–West pipeline 4–5 million barrels a day in recent months, 4–5% of global supply; capacity 7 million barrels a day HVG, 12 September 2026; Portfolio, 12 September 2026
Saudi crude exports, August 2026 around 3 million barrels a day, the lowest since 2017 Portfolio, 12 September 2026
Crude flows through the Strait of Hormuz from 4 million barrels a day before the war to 1 million Portfolio, 12 September 2026
Brent price close to 110 dollars on Thursday, 104 dollars/barrel on Friday; weekly rise of more than 8% Magyar Nemzet, 12 September 2026
Price level compared with before the war crude oil +53%, European diesel +89% Mol, 8 September 2026, as reported by Magyar Nemzet
Weight of Bab el-Mandeb 10–12% of global seaborne trade 24.hu, 12 September 2026
Diesel support HUF 5,000 a month for diesel passenger cars of up to 150 horsepower (110 kW), paid automatically by the Treasury; nearly one million eligible according to the government ATV, 13 September 2026
Budgetary burden of the diesel support around HUF 6 billion HVG, 12 September 2026
Road-haulage sector nearly 11,600 hauliers, 4% of gross domestic product (GDP), more than half of domestic diesel consumption NiT Hungary, as reported by HVG

Two data points require a separate note. The REKK article gives the lost volume as “922 thousand gigawatt-hours”; this is an obvious unit error, because Hungary’s total annual electricity consumption is a fraction of that. With the result of 37.4 billion forints and the production cost of 14 forints, an order of magnitude of 922 gigawatt-hours, i.e. 922 thousand megawatt-hours, is consistent — this implies an average replacement price of around 55 forints per kilowatt-hour. The sources give differing figures for the total amount of the diesel support (the product of the monthly amount and the duration does not match the total reported), so this analysis uses only the monthly amount.

6.3 Policy dimensions

  • Economy (programme points) — cost-bearing and shock preparedness: G25 (energy-price shock preparedness plan) provides the frame for the severe scenario and the shock-cost statement, G19 (radical transparency in economic decision-making) the publicity of the report, G1 (data-driven budget) the named budget line; the countercyclical fiscal stabiliser under the G15 programme point — public spending in a downturn, saving in an upswing — the logic of funding shock expenditure;
  • Environment and climate (programme points) — the automatic thresholds of K7 (energy-market shock resilience) and the reduction of exposure under K2 (energy transition plan): capacity dependent on cooling water and capacity dependent on fossil imports both carry climate and geopolitical risk;
  • Foreign policy (programme points) — within KP13 (geopolitical situation-analysis capacity), regular assessment of Bab el-Mandeb as a European trade and data-cable chokepoint;
  • Transport and infrastructure (background material) — the diesel exposure of the road-haulage sector and the possibility of a liquidity instrument.

6.4 Literature in detail

6.4.1 Adam Smith: The Wealth of Nations

In Book V Smith compares joint-stock companies with private copartneries. He describes how most proprietors neither understand nor follow the company’s affairs, content themselves with the dividend fixed by the directors, and the company is run by a court of directors. From this he derives one of the most quoted observations of the work:

“The directors of such companies, however, being the managers rather of other people’s money than of their own, it cannot well be expected that they should watch over it with the same anxious vigilance with which the partners in a private copartnery frequently watch over their own. […] Negligence and profusion, therefore, must always prevail, more or less, in the management of the affairs of such a company.”

Smith adds that such companies have seldom been able to compete without an exclusive privilege — that is, weak vigilance was often made sustainable by a protected market position.

Applied to the Paks case, the thesis is not a personal verdict on MVM’s managers but an incentive diagnosis. Under state ownership “other people’s money” belongs to an even more distant owner: the community, which does not directly follow the company’s risk management. If the damage from the low-water risk can later migrate to the budget or the consumer, then failing to make the risk-reducing investment carries no price at the company — by Smith’s logic, neglect is then not an exception but expected behaviour. The cost-bearing rule under 3.2 replaces this missing ownership vigilance with a consequence known in advance.

📖 Source: Adam Smith: The Wealth of Nations

6.4.2 IMF: World Economic Outlook 2024

In its special commodities section, the IMF’s October 2024 report recorded that fears of escalation in the Middle East had built a volatile risk premium into oil prices, and that “a rise in Red Sea maritime attacks has dislocated seaborne oil flows, decreasing traffic through the Suez Canal by almost two-thirds”, largely rerouting shipments around the Cape of Good Hope. The report’s opening chapter also makes a more general statement:

“We have now entered a world dominated by supply disruptions–from climate, health, and geopolitics.”

The report also shows that oil supply shocks mainly raise headline rather than core inflation, which is why central banks have traditionally “looked through” such shocks if they are not too large. In Hungary the interest-rate decision belongs to the Monetary Council of the Hungarian National Bank (MNB); fiscal preparedness, however, is the responsibility of the government and cannot wait for the monetary reaction.

Two lessons follow for the present situation. One is that the vulnerability of the Red Sea route is not new information: it has been documented since the 2024 precedent, so treating the severe scenario as the planning baseline is not over-insurance but the incorporation of existing knowledge. The other is that the report puts supply disruptions of climate and geopolitical origin in a single category — the low water at Paks and the shutdown of the Saudi pipeline require the same preparedness logic, which justifies a joint shock-cost statement.

📖 Source: IMF: World Economic Outlook 2024 — Policy Pivot, Rising Threats

6.4.3 Ray Dalio: Principles

At the centre of Dalio’s principles of organisational management lies the handling of mistakes. In his view most learning comes from making mistakes and reflecting on them objectively, which he condenses into the formula that pain plus reflection equals progress. Two of his principles are particularly relevant. One is radical transparency: those concerned should be given the most direct possible view of what is happening around them, because direct access allows them to form their own opinions, and this improves the accuracy of decisions. The other is the distinction between proximate causes and root causes: the proximate cause is the action or omission that led to the problem, while the root cause is the deeper characteristic that produced the omission — and the problem is eliminated only by removing the root cause. Dalio also warns that organisations usually avoid examining mistakes because it is painful, so the objective diagnosis does not happen.

The limit of transferring this is obvious: Dalio writes about the internal culture of an investment firm, not a public institution, and in the public sector national-security or nuclear-safety considerations may also justify classification. In the case of the Paks report, however, it is this frame that provides the strongest argument for publicity. The proximate cause — the Danube’s low water level — is known to everyone; the root-cause question, by contrast, is why a risk recorded in official reports did not become an investment decision. If the report remains an internal document, it is precisely this diagnosis that does not happen, and at the next outage the same root cause keeps operating.

📖 Source: Ray Dalio: Principles

6.5 International comparison

The closest established model for a cost-bearing rule is the so-called prudence review in US utility regulation: the state regulatory authority examines the costs that a utility wishes to build into its tariff against the yardstick of whether a prudent operator could have avoided them on the basis of the information known at the time of the decision. Costs arising from an imprudent decision cannot be built into consumer prices; they are borne by the company’s owners. MIAK’s proposal under 3.2 is a simplified version adapted to the Hungarian situation: instead of “information known at the time of the decision”, the criterion is a risk recorded in a public official document, because this can be verified afterwards without dispute.

Among state-owned nuclear operators, France’s EDF faced a similar question during the mass reactor shutdowns caused by corrosion problems in 2022: the company reported the financial impact of the lost output in its public financial statements, and the deterioration of its financial position also played a part in the state later acquiring full ownership of the company. The lesson for the present topic is that state ownership does not make the loss invisible — at most it decides whether it appears in a public annual report or quietly in the budget.

On the oil side, Orlen’s statement of 12 September is the most direct counter-example: the company told the PAP news agency that the shutdown of the Saudi pipeline does not affect its refineries’ crude supply because it consistently diversifies its sources; in 2026 a Norwegian contract may cover a quarter of its crude needs, and it has also expanded its own production share. According to the Hungarian company Mol’s statement of 8 September, domestic physical supply is not at risk in the short term either, because shipments largely arrive via a Croatian port and pipeline — the risk here arises in pricing, not in volume. This strengthens the logic of proposal 3.3: the primary field of the Hungarian shock response is managing costs and allocating them in advance.

Economy

  • G1 — Data-driven budget
  • G15 — Countercyclical fiscal stabiliser
  • G19 — Radical transparency in economic decision-making
  • G25 — Energy-price shock preparedness plan

Environment and climate

  • K2 — Energy transition plan
  • K7 — Energy-market shock resilience

Foreign policy

  • KP13 — Geopolitical situation-analysis capacity

Proposed new programme point: Outage cost-bearing rule for state-owned energy companies — for the Economy policy area: additional cost arising from the outage of a strategic generating unit is charged to the owner if the risk previously appeared in a public official document; for an unforeseeable risk, socialisation may take place only on a named budget line, without building it into prices; with an annually updated public shock-cost statement.

6.7 List of sources

Press sources (MIAK press monitor, 13 September 2026 — topic 2):

Knowledge-base references (books):

  • 📖 Adam Smith: The Wealth of Nations
  • 📖 IMF: World Economic Outlook 2024 — Policy Pivot, Rising Threats
  • 📖 Ray Dalio: Principles

MIAK internal materials:

  • MIAK policy area: Economy (programme points; programme point ID: G1, G15, G19, G25)
  • MIAK policy area: Environment and climate (programme points; programme point ID: K2, K7)
  • MIAK policy area: Foreign policy (programme points; programme point ID: KP13)
  • MIAK policy area: Transport and infrastructure (background material)
  • MIAK press monitor, 13 September 2026 — topic 2, score: 93/100
  • MIAK international press monitor, 13 September 2026 — topic 3 (background)

Supplementary public data sources:

  • ENTSO-E Transparency Platform — Hungarian quarter-hourly wholesale electricity prices (the data series identical to HUPX power exchange prices)
  • MAVIR — nuclear power plant generation data
  • Hungarian Atomic Energy Authority — 2025 national report under the Convention on Nuclear Safety
  • Annual reports of the Paks Nuclear Power Plant and MVM
  • IEA Oil Market Report; Eurostat Weekly Oil Bulletin

Generation metadata