Part I — Situation overview

On 5 August 2026, on the first day of the two-day cabinet meeting, the government adopted a comprehensive energy development plan. The Prime Minister set out the decision in the evening on his social media page: the framework sum of the first phase is 868 billion forints, and the use of the funds is divided in four directions — energy storage capacity, modernisation of the electricity network, wind energy and geothermal investments. Financing will be realised from domestic and EU sources, with the involvement of private capital. The announcement also contained several concrete points: the network connection of household solar panels will be accelerated and made more transparent, the licensing of geothermal energy exploration will be simplified, and the possibility of requesting free smart meters will be introduced. The smart meter will be compulsory for households consuming more than 4,000 kilowatt hours a year. In the wind power tenders to be issued by the end of August a mandatory state shareholding will be prescribed, so that the financial benefit of the developments should go to the budget and to local municipalities; settlements can apply for participation within the frame of the Saint Stephen Rural Development Programme. On Friday, moreover, the result will also be announced of a network development tender worth one and a half billion euros — close to 600 billion forints — financed from EU sources.

The framing of the announcement was unambiguous. The Prime Minister traced the present vulnerability back to the decisions of his predecessors: “The construction of wind farms in Hungary was deliberately obstructed. The development of storage capacities was neglected. […] And Paks was not prepared for the low water levels now being experienced” — he put it, adding that the previous government had also renounced the EU funds available for energy developments and climate adaptation. The background is indeed pressing: alongside the record temperature of 43 degrees Celsius measured in Budapest, the derating of the Paks Nuclear Power Plant on Tuesday was 1,770 megawatts (MW), that is, the plant was generating at 230 megawatts instead of 2,000. Gross peak system load meanwhile reached 6,800 megawatts — 300 megawatts less than the planned 7,100 megawatts. Alongside the decision the Prime Minister also gave instruction that every policy area should prepare its medium- and long-term adaptation strategy.

By MIAK’s reading the content of the programme is good, but its decision procedure is incomplete — and the two are not the same question. The package answers precisely those gaps that the MIAK programme has named for two years: storage capacity, network flexibility, demand-side efficiency. Not a single element seems unjustified. The trouble is that a distribution of 868 billion forints today carries no public, pre-fixed output indicator, no public impact assessment, and no public economic justification for its single most contestable element, the mandatory state shareholding either. This is exactly the decision pattern — a large sum, in a crisis, without justification — that MIAK regularly objected to with the previous government. The yardstick does not become less valid because it is now somebody else who distributes the funds.

Part II — Literature foundation

Before turning to MIAK’s proposals it is worth setting out the conceptual frame in which the fact that the professional content of a large programme may be right while its decision procedure is nevertheless risky can be read. The central proposition of The Effective Executive by Peter F. Drucker (Austrian-American management thinker, founder of the methodology of organisational efficiency measurement) is that effectiveness rests not on the quantity of work but on the discipline of priority decisions: concentration on a few key areas and the deliberate abandonment of programmes that do not work are alike preconditions of measurement. The volume Gyors és lassú gondolkodás (Thinking, Fast and Slow) by Daniel Kahneman (Israeli-American cognitive psychologist, laureate of the Nobel memorial prize in economics) gives the other side: the concept of the planning fallacy describes the systematic bias under whose effect the implementers of large programmes overestimate benefits, underestimate costs and treat the best case as the base plan — and the author also presents a concrete antidote to this. And the work The Innovator’s Dilemma by Clayton M. Christensen (American business thinker, creator of the theory of disruptive innovation) explains why storage and flexible capacity did not receive funding in a system optimised for baseload plants: it was not the intention that was missing, but the resource allocation process was working logically by its own yardstick. The detailed treatment of the literature — author by author, with quotations — can be found in the 6.4 Literature in detail section.

📖 Source: Peter F. Drucker: The Effective Executive; Daniel Kahneman: Gyors és lassú gondolkodás (Thinking, Fast and Slow); Clayton M. Christensen: The Innovator’s Dilemma

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. None of them disputes the content of the programme, and none of them asks for new funds: all three are about the decision already taken being evaluable afterwards.

3.1 Publication of output indicators by sub-programme (at the same time as the tenders are issued)

The 868 billion forints today appears under four collective headings — storage, network, wind energy, geothermal — without output indicators by sub-programme. MIAK proposes that at the same time as the wind power tenders are issued at the end of August, two or three pre-fixed, quantified indicators should be published for each sub-programme, against which delivery can be measured afterwards: for storage, the capacity to be installed in megawatts and the discharge duration; for the network, the improvement of N-1 operational security compliance and the newly connectable solar capacity; for wind energy, the megawatts expected from the tenders and the date of first commissioning; and for geothermal, the reduction of licensing lead time in days. These are not new calculations: a tender is not drawn up without the technical estimate behind it — the proposal amounts only to that estimate also being placed before the public, and not only in the decision preparation material. The reason is simple: if there is no expected result stated in advance, then in two years’ time the programme will be neither successful nor unsuccessful, only spent. The proposal follows from the G19 radical transparency and the D2 open data programme points, and in its measurement logic builds on the KI8 efficiency measurement. In Drucker’s frame (see 6.4.1) this is the step that turns intention into a verifiable commitment.

3.2 Reference class based cost and deadline estimation for the large sub-programmes (before contracting)

The typical failure of large infrastructure programmes is not the wrong goal but the regular overrun of cost and deadline. MIAK proposes that for every sub-programme of the programme above 50 billion forints a short, public estimate check should be prepared before contracting, according to the method developed by Kahneman and Bent Flyvbjerg: the estimate should be derived not only from the given project’s own plan but also from the actual cost and deadline data of similar, already completed domestic and European projects, and the difference between the two should be described in a single paragraph. If the own estimate is significantly below the average of similar projects, that has to be justified. This step runs to a few pages, yet catches exactly the bias because of which large programmes prove more expensive afterwards — and because of which the budget reserve regularly runs out. The G20 impact assessment system prescribes an ex post audit; this proposal is the forward-looking counterpart of the same. The mechanism of the planning fallacy is expounded in section 6.4.2.

3.3 Public economic justification of the mandatory state shareholding (as an annex to the tender documentation)

The single most contestable element of the programme is the mandatory state shareholding written into the wind power tenders. This is an industrial policy decision in its own right, for which there are arguments on both sides: for it, keeping the rent at home and the local revenue; against it, the rise in the cost of capital and the narrowing of the pool of bidders. MIAK does not ask the government to abandon it — it asks that a short, public analysis on four questions should appear as an annex to the tender documentation: how large a shareholding is at stake, how it affects the cost of capital of the projects and thereby the capacity actually realised, what competitive effect the government expects on the composition of the pool of bidders, and what corporate governance rule ensures that the state shareholding does not become a channel of subsequent intervention. The fourth question is the most important: ownership powers and regulatory powers are two different things, and if the same actor exercises both, the conflict of interest has to be handled with separate guarantees. The proposal stands at the intersection of the G9 strategic industrial policy and the G5 competition policy programme points, and follows the G6 logic against regulatory capture.

The three proposals can be strung on a single principle: a large sum is not a legitimation but an obligation. An 868 billion forint programme is the more defensible the more precisely it is written down in advance what we expect from it — because then the debate afterwards is about delivery, not about intentions. Christensen’s argument (see 6.4.3) adds that without changing the yardstick the resource, even with good intentions, flows away in the habitual, easily calculable direction — that is, the designation of indicators is not an administrative extra but the only instrument by which the structure of the programme can be turned towards the bottleneck.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Economy Pre-fixed output indicators make the yield of the programme priceable and provide the basis for sizing the next phase A public indicator gives a political attack surface, which may breed cautious, low-set target figures
Society The local shareholding and the free smart meter are directly perceptible benefits, which strengthen acceptance of the programme If the compulsory part of the smart meter later comes with a change in price structure, that may cause a loss of trust without prior information
Public administration Reference class based estimate checking takes large programmes out of the routine of optimistic planning The extra step slows the issuing of tenders, which is a real cost given the crisis timing

The main question to be weighed lies between speed and verifiability. In the middle of a crisis every day counts, and every prior procedural step — impact assessment, consultation, estimate checking — takes time. This is a real cost, not a formality. MIAK nevertheless argues for the prior steps, for two reasons. First: of the programme’s four directions not a single element starts tomorrow, the wind power tenders are issued at the end of August, and the EU tender for network development has a lead time of several years — that is, the time required for writing the indicators is taken not from crisis management but from planning. Second: a programme without indicators is not faster, only less verifiable; delay comes to light afterwards, whereas an unfulfilled goal that was never even stated does not.

The proposal tips over to the risk side if writing the indicators becomes a genre of its own. If twenty indicators per sub-programme end up in the tender, then it will not be measurement but administration, and the energy of implementation will go into report writing. This is why MIAK proposes two or three indicators per sub-programme and a half-yearly report, no more: measurement works if it is sparse, if it is public, and if it has a consequence. And the greatest silent risk is not in the internal structure of the programme but in its timing: the smart meter requirement and the mandatory state shareholding are two elements that fix the structure of the market for years, and are being decided now, on the tenth day of a heatwave.

Part V — Measurability and summary

5.1 What is worth tracking? (proposed KPIs)

Four performance indicators (KPIs) are worth tracking over the coming eighteen months:

  1. Contracted storage capacity: the electricity storage capacity contracted from the programme in megawatts, in quarterly breakdown — proposed target: a measurable, publicly tracked path up to mid-2027, not a single closing figure.
  2. Smart meter penetration: the share of households equipped with a smart meter — proposed target: an annually published growth path starting from the present 11 per cent level, which is low in European comparison.
  3. Effectiveness of the wind power tenders: the ratio of capacity tendered to capacity actually contracted, and the number of bidders per tender — proposed target: at least three valid bids per tender, because below this the competition-narrowing effect of the mandatory state shareholding is confirmed.
  4. Estimate accuracy: the actual cost of closed sub-programmes as a percentage of the original estimate — proposed target: the overrun should remain durably below 15 per cent, and the value should appear for every closed sub-programme.

5.2 Summary

MIAK’s key message in a single sentence: the professional content of the 868 billion programme is defensible, but its decision procedure is today not verifiable, and this may afterwards render the programme indefensible as well. Concretely, it asks the government to publish the output indicators of the sub-programmes at the same time as the wind power tenders are issued at the end of August, to carry out for sub-programmes above 50 billion forints, before contracting, the estimate check based on the actual data of similar projects, and to attach a public economic justification to the mandatory state shareholding. And it asks the public to measure the programme not by its sum but by its committed numbers — the sum is an announcement, the number is a commitment.

Two MIAK foundational values are in play in this matter. Data-drivenness, because the quality of an allocation decision taken in a crisis becomes judgeable only afterwards, and only if there is something to measure it against: without a figure fixed in advance, delivery of the programme and failure of the programme will look exactly the same. And accountability, because the justification of the programme at present rests on a claim about the past — the omissions of the predecessors — which may perhaps be sound in a political sense, but in an accountability sense is empty: an enumeration of the developments not made does not tell us whether the present 868 billion is well allocated. Accountability begins with the measurement of one’s own commitment, not with an inventory of the previous actor’s mistakes.

Part VI — Justifications and further sources

6.1 The press framing by spectrum

The economic band gave the only substantive treatment of the programme. Portfolio carried the news in two separate pieces: one the bare fact of the decision and the four development directions, the other — the detailed summary — the concrete points, the 4,000 kilowatt hour threshold for smart meters, the role of the Saint Stephen Rural Development Programme, and the network development tender worth one and a half billion euros whose result is to be announced on Friday. The same paper also gave the most uncomfortable analysis of the day, on the point that always-available, secure and greener energy cannot be durably cheap — that is, the price structure question behind the programme was posed by the economic band alone, even if not in relation to the programme.

The left-liberal band sharpened the news towards the political framing of the announcement. 444.hu raised the Prime Minister’s responsibility-attributing sentences into the headline and the body text, quoting verbatim the claims concerning his predecessors; Telex put the single most concrete element of the programme, the issuing of the wind power tenders and the state shareholding prescribed in them, at the centre of its headline; HVG carried the fact of the announcement. This band therefore worked out the political reading of the decision, not its structure.

The conservative band made two mutually complementary moves. Mandiner’s report recorded drily that the Prime Minister “did not, however, give concrete details” — this is the most precise press observation of the day on the information content of the announcement — and then also quoted the prime ministerial sentence on national unity presupposing international recognition. Magyar Nemzet, by contrast, took the topic to the functioning of the responsible ministry, putting a leadership crisis and a ministry occupied with other things instead of crisis management at the centre of its headline. The public affairs band brought the effects side: ATV the summary of the announcement, and 24.hu — still before the decision — the system-level role of energy storage. One framing was missing from every band: not a single paper asked on the basis of what indicator it will be judgeable whether the 868 billion forints reached its goal.

6.2 Facts and data

Datum Value Source
Framework sum of the first phase of the energy development programme HUF 868 billion Prime ministerial announcement, Portfolio, 5 August 2026
Date of the decision 5 August 2026, the first day of the two-day cabinet meeting Portfolio, 5 August 2026
Development directions energy storage, network modernisation, wind energy, geothermal Portfolio, Mandiner, 5 August 2026
Financing source domestic and EU funds, with the involvement of private capital Mandiner, 5 August 2026
Threshold of the smart meter obligation household consumption above 4,000 kWh a year Portfolio, 5 August 2026
Share of households equipped with a smart meter 11% Euractiv, 6 August 2026
Deadline for issuing the wind power tenders end of August 2026 Portfolio, 5 August 2026
Size of the EU network development tender EUR 1.5 billion (close to HUF 600 billion) Portfolio, 5 August 2026
Derating of the Paks Nuclear Power Plant (4 August, Tuesday) 1,770 MW — generating at 230 MW instead of 2,000 MW MAVIR data, kormany.hu, Mandiner, 5 August 2026
Gross peak system load (Tuesday) 6,800 MW MAVIR data, kormany.hu, Mandiner, 5 August 2026
Pre-planned peak system load (Tuesday) 7,100 MW MAVIR data, kormany.hu, Mandiner, 5 August 2026
Daily temperature record in Budapest 43 °C Portfolio, 5 August 2026
The water level of the Danube at Paks on the day of the announcement exceeding the minimum level by 7 cm Portfolio, 5 August 2026

A single connection deserves to be highlighted. Tuesday’s figure — an actual peak load of 6,800 megawatts against the planned 7,100 — shows that demand-side intervention worked measurably: the communication on rescheduling consumption and the voluntary restraint of large consumers were worth 300 megawatts. This figure is comparable in order of magnitude with what the storage branch of the programme will deliver over years — yet from the demand side only the smart meter appears in the 868 billion package, with no renovation programme. On this day another, non-governmental initiative gave the demand-side answer: the member companies of the Association of Real Estate Developers Roundtable worked out a voluntary package of professional recommendations for the operation of office buildings and logistics properties — review of cooling and lighting systems, rescheduling of electric car charging outside the peak between 5 and 10 p.m., temporary switching off of façade and advertising lighting. These are not rules that have entered into force but industry self-regulation; and precisely for that reason it is worth noting that the largest immediate reserve this summer came not from the government programme but from the demand side.

6.3 Policy dimensions

  • Environment and climate (programme points) — the four branches of the programme are directly the content of the energy transition plan and of shock resilience; storage and network flexibility are here the dispatchable pillar (programme point ID: K2, K7, K1);
  • Economy (programme points) — the intersection of the impact assessment system, strategic industrial policy and competition policy: the mandatory state shareholding opens all three questions at once (programme point ID: G20, G9, G5, G6, G19, G10);
  • Construction (programme points) — the missing demand-side branch: the renovation of the building stock is the largest untapped reserve (programme point ID: EP3, K6);
  • Public administration and e-government (programme points) — the institutionalisation of large-programme monitoring and of behavioural intervention (programme point ID: KI8, KI5);
  • Digitalisation and AI regulation (programme points) — the machine-readable publication of sub-programme indicators and programme progress (programme point ID: D2).

6.4 Literature in detail

6.4.1 Peter F. Drucker: The Effective Executive

The central claim of Drucker’s volume is that executive effectiveness rests not on the quantity of work but on the discipline of priority decisions. In the author’s formulation, effective executives “concentrate on the few major areas where superior performance will produce outstanding results”, and force themselves to stick to their priority decisions. The edge of the argument, however, is not concentration but its converse: according to the author the executive

“has no choice but to do first things first — and second things not at all.”

Drucker attaches two further elements to this. One is organised abandonment: by his account the first rule of concentration is that the past which no longer produces has to be wound down. The other is the methodology of decision-making: a valid decision is based not on agreement about the facts but on the clash of dissenting opinions.

In the case of the 868 billion programme this means that the present structure of the package — four directions, without a priority order and output indicators by sub-programme — is exactly the form that Drucker would not regard as a priority decision: an enumeration is not a ranking. Proposal 3.1 therefore asks not for more data but for less and sharper: two or three numbers per sub-programme against which the programme can be measured. And the requirement of dissenting opinions supports proposal 3.3: the mandatory state shareholding is precisely the element where the public counter-argument is not a brake on the decision but the condition of its quality.

📖 Source: Peter F. Drucker: The Effective Executive

6.4.2 Daniel Kahneman: Gyors és lassú gondolkodás (Thinking, Fast and Slow)

Kahneman introduced the concept of the planning fallacy for the systematic bias under whose effect plans and forecasts stand unrealistically close to the optimal case. By his analysis the implementers of risky programmes “overestimate benefits and underestimate costs”, and draw up the scenarios of success while overlooking the possibility of errors and miscalculations. As an illustration of this the author cites a 2002 survey in which American homeowners renovating their kitchens estimated the cost on average at 18,658 dollars — and paid on average 38,769 dollars.

The strength of the book, however, is not the diagnosis but the procedure offered with it. Kahneman sets out the method of Danish planning expert Bent Flyvbjerg, which consists of three steps: identify a reference class of similar, already completed projects; prepare a baseline forecast from their actual cost and deadline data; and then correct that baseline estimate in the light of knowledge of the individual case. The author also records that decision-makers, “before finally accepting the plan in question”, have to assess realistically the costs and benefits of the offer — the budget reserve is in itself no protection, because, in the formulation of an official quoted in the book which was given to Flyvbjerg, it is to the contractors like raw meat to lions.

For the Hungarian programme this is directly applicable. For the storage and network sub-programmes there is ample reference: the European battery storage projects and network investments of recent years have cost and lead time data per megawatt. Proposal 3.2 asks precisely for this — not for another genre of impact assessment, but for a few pages on where the Hungarian estimate stands compared with the actual data of similar projects, and, if significantly below them, why.

📖 Source: Daniel Kahneman: Gyors és lassú gondolkodás (Thinking, Fast and Slow)

6.4.3 Clayton M. Christensen: The Innovator’s Dilemma

The explanatory power of Christensen’s volume lies in the fact that it traces the failure of incumbent players not to error but to rationality. By his argument the resource allocation process — the mechanism that decides which initiative gets people and money — necessarily pushes investment towards the larger and more certain return. In the author’s formulation

“in the battle for development resources, projects targeted at the articulated needs of current customers always win over proposals to develop products for markets that do not exist.”

The volume adds that disruptive products are typically simpler and promise lower margins, so the most profitable customers do not want them at first — and that it is precisely the well-working, “sensible” resource allocation process that filters out such proposals.

The development history of the Hungarian electricity system is an example of this mechanism, not a simple series of omissions. In a system optimised for baseload plants and solar build-out, both directions offered a readily calculable return, whereas storage and dispatchable flexible capacity had a longer payback and were more uncertain — so in every single decision round they logically lost out. MIAK draws two conclusions from this. One is that responsibility is distributed across several government cycles, and the part of the present announcement relating to the past therefore does not bring us closer to a solution. The other is that without changing the yardstick, the 868 billion now allocated will also flow away in the habitual direction: if the delivery of the programme is measured in installed capacity, then installed capacity will grow; if in the reserve available in the hardest evening hour, then the reserve will.

📖 Source: Clayton M. Christensen: The Innovator’s Dilemma — When New Technologies Cause Great Firms to Fail

6.5 International comparison

The programme is worth reading in the regional frame in which it was born. In Romania the government declared a national energy alert on 31 July for the whole of August, and then more than eighty energy-intensive large companies — among them car factories, steel and fertiliser producers — began maintenance or output reduction; Dacia and Ford halted production until 19 August, voluntarily cutting consumption by around 200 megawatts. In Serbia the record low level of the Danube has held back the generation of the Đerdap hydropower plants and is also obstructing river fuel imports. That is, the Hungarian programme is not a unique domestic step but one element of a regionally parallel series of responses — and this parallelism is in itself an argument for measurability: the investment decisions of neighbouring systems will be comparable over the coming years, if there is anything to compare.

The methodological model, however, is worth taking not from the region but from European and British public investment practice. Reference class based cost estimation — which Kahneman sets out from Flyvbjerg’s work — is not a theoretical proposal: it is an established element in the preparation of large transport and infrastructure programmes, precisely because the actual data of similar projects are the strongest constraint on optimistic planning. In the case of an 868 billion programme running over several years the cost of this step is a few expert weeks; its benefit is that in two years’ time the programme will provoke debate not about its intentions but about its delivery.

Environment and climate

  • K2 — Energy transition plan
  • K7 — Energy market shock resilience
  • K1 — Measurable climate targets
  • K6 — Building energy efficiency programme

Economy

  • G20 — Economic policy impact assessment system (Drucker audit)
  • G19 — Radical transparency in economic decision-making
  • G9 — Strategic industrial policy
  • G5 — Competition policy and anti-monopoly
  • G6 — Programme against rent-seeking and regulatory capture
  • G10 — State development bank
  • G25 — Energy price shock preparedness plan

Construction

  • EP3 — Energy efficiency renovation programme

Public administration and e-government

  • KI8 — Drucker-style efficiency measurement in public administration
  • KI5 — Behavioural public policy unit (Nudge Unit)

Digitalisation and AI regulation

  • D2 — Open data programme

Proposed new programme point: Reference class based cost and deadline estimation for state investment sub-programmes above 50 billion forints — for the Economy area, as the forward-looking counterpart of the G20 impact assessment system.

6.7 List of sources

Press sources (MIAK press monitor, 6 August 2026 — topic 1):

Knowledge-base references (books):

  • 📖 Peter F. Drucker: The Effective Executive
  • 📖 Daniel Kahneman: Gyors és lassú gondolkodás (Thinking, Fast and Slow)
  • 📖 Clayton M. Christensen: The Innovator’s Dilemma — When New Technologies Cause Great Firms to Fail

Note: the local file path of the books does NOT appear in the visible text of the blog — only the author and the title. The file path is an internal matter of the generation process, not the reader’s.

MIAK internal materials:

  • MIAK policy area: Environment and climate (programme points; programme point ID: K2, K7, K1)
  • MIAK policy area: Economy (programme points; programme point ID: G20, G9, G5, G6)
  • MIAK policy area: Construction (programme points; programme point ID: EP3)
  • MIAK policy area: Public administration and e-government (programme points; programme point ID: KI8, KI5)
  • MIAK press monitor, 6 August 2026 — topic 1, score: 95/100

Additional public data sources:

  • MAVIR — daily system load and schedule data
  • kormany.hu — the daily reports prepared on the heat alert
  • ENTSO-E Transparency Platform — cross-border capacity and physical flows
  • National Directorate of Water Management — Danube water level time series, Paks

Generation metadata