Part I — Situation overview

On Monday, 19 October, Prime Minister Péter Magyar presented the main figures of the 2027 budget to the National Assembly in a pre-agenda speech of more than two hours. The budget deficit target (the annual general government deficit fixed in advance as a share of gross domestic product, i.e. GDP) is 4.9 per cent. On Tuesday, on Kossuth Radio, the head of government spoke of a 2.4 per cent growth assumption and a 3 per cent inflation assumption, and of the need to reduce the deficit to 2.8 per cent by 2030 (Telex). According to Portfolio’s tally, the government decided on roughly HUF 2,100 billion of deficit-increasing measures. Personal income tax (SZJA) on the minimum wage falls to 9 per cent, everyone’s tax falls up to a gross monthly income of HUF 680 thousand, the family allowance doubles, and more than 600 thousand public-service employees get a pay rise. Against this stands an improving package of more than HUF 3,000 billion. More than two-thirds of it is spending cuts: maintaining the 700 billion saving achieved this year and HUF 1,500 billion of new cuts. On the revenue side comes the wealth tax, and the corporate tax rises for companies with revenue above HUF 180 billion.

At the same sitting, the National Assembly adopted the amendment of the 2026 budget with 134 votes in favour, 48 against and one abstention. The deficit target changed from 3.7 to 7.5 per cent, the growth assumption from 4.1 to 2 per cent, and government debt as a share of GDP may rise from last year’s 74.6 per cent to 77.5 per cent (Telex). In the debate, the president of the Fiscal Council, Gábor Horváth, recalled that this is the eighth year since 2015 in which the budget has had to be amended during the year (444.hu). MIAK’s analysis of yesterday, written before the presentation, dealt with the end point of the deficit path, the inflationary effect of the utility subsidy and official tax-burden statistics. This piece does not repeat them.

Most analysts consider 4.9 per cent achievable, but almost all of them point to the same spot. According to Portfolio, of the 2.6 percentage-point (pp) improvement, 1.4 points would occur even without measures, as the one-off effects of this election year run out. The net effect of the government’s own measures is thus roughly HUF 1,000 billion. According to Gábor Regős, asked by Index, the success of the budget depends on whether the 2,200-billion saving materialises. MIAK’s reading: the credibility of the 2027 budget is decided in three places. The first is whether the saving appears in concrete appropriations. The second is whether an error in the growth assumption is priced in advance. The third is whether the revenue from the corporate tax rise shows the amount that can actually be collected.

Part II — Grounding in the literature

Three sources provide the framework for interpreting the three questions. Peter F. Drucker (Austrian-American management thinker, founder of the methodology of measuring organisational effectiveness) writes in his book The Effective Executive that feedback must be built into every decision from the outset: the expectations behind the decision must be continuously tested against actual events. A “savings target” is therefore a decision only if it breaks down into items fixed in advance that can be held to account afterwards. A box in the International Monetary Fund’s (IMF) 2023 World Economic Outlook report showed that after the 2008 crisis medium-term growth forecasts were biased upwards rather than downwards. Actual growth regularly fell short of expectations. The 4.1 per cent assumption of Hungary’s 2026 budget is a textbook case of this. János Kornai (Hungarian-born economist, author of the theory of the soft budget constraint; Harvard professor from 1986 to 2002) distinguishes between hard and soft taxation in his work Economics of Shortage (A hiány). In soft taxation, the company influences the rules and can obtain allowances through individual exemptions. “Targeting” large-company tax allowances moves exactly along this boundary. The detailed discussion of the literature — author by author, with a paraphrase of the line of reasoning and a quotation — is in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. All three relate to the bill to be submitted at the end of the month, and all three aim to turn the figures announced on Monday into verifiable appropriations.

3.1 The 2,200-billion saving chapter by chapter in the bill, with the release of the 300-billion reserve tied to mid-year delivery (with the submission, by the end of October; first check by 31 July 2027)

Monday’s announcement stated the spending saving as a single sum: 700 billion from maintaining this year’s measures, 1,500 billion from new cuts. The Prime Minister mentioned overpriced procurement, concessions, government communication, the public media and the public-interest asset-management foundations (kekva). According to Erste’s analyst, Orsolya Nyeste, the saving expected from curbing waste and corruption, amounting to about 1.5 per cent of GDP, carries implementation risk. MIAK proposes that the bill submitted by the Government should not show this amount on a central, unallocated line, but as reductions in chapter appropriations, by ministry and by institution. It should be accompanied by a measure-by-measure register. For each item it should state the baseline amount, the expected saving and the type: terminated contract, renegotiated price, operating expenditure or postponed investment. The last is not a saving, only a shift in time, so it must be marked separately. According to Index, the plan also contains a HUF 300 billion safety reserve. MIAK would tie its release to at least half of the chapter savings having been delivered according to the data for the first half of 2027. Drucker’s feedback principle (see 6.4.1) asks for exactly this: fix the expectation at the time of the decision and measure it against reality. The proposal links the G21 spending review and the KI8 efficiency measurement. On 25 August MIAK asked for the itemised publication of this year’s 700 billion, and on 14 October for the identification ratio of the adjustment. This proposal goes one step further: the saving should not only be known, but locked in by law.

3.2 A sensitivity annex to the 2.4 per cent growth assumption and an in-year deviation threshold (with the submission; the threshold rule by the final vote)

The 2026 budget assumed 4.1 per cent growth; the amended law assumes only 2 per cent. Telex recalls that at the time of the original planning independent analysts expected 1–2 per cent. The 2027 assumption of 2.4 per cent is more cautious and roughly in line with market estimates. According to K&H’s analyst, Dávid Németh, however, the deficit could be larger if the government’s macroeconomic calculation does not work out. MIAK proposes two steps. First: the bill should get a sensitivity annex. It should show how much the deficit changes if growth falls short by 1 percentage point, if inflation deviates by 1 percentage point, and if government-bond yields rise by 1 percentage point. Second: an in-year deviation threshold should be put into the law. If, according to the Ministry of Finance’s quarterly forecast, the deficit exceeds the target by more than 0.3 per cent of GDP, the Government should submit an offsetting proposal within 60 days. It should not settle the difference at the end of the year with an ex-post amendment. The Fiscal Council should give a public opinion on it. According to the IMF’s 2023 analysis (see 6.4.2), medium-term growth forecasts tend towards excessive optimism. The error should therefore not be ruled out, but priced in advance. The proposal is an element of the G1 data-driven budget and the G23 debt-sustainability framework. On 6 October MIAK proposed an independent path forecast by the Fiscal Council. This proposal sets out what should happen if the assumption collapses during the year.

3.3 Corporate-tax rise: gross and net revenue together with the global minimum tax, and a public set of conditions for targeted tax allowances (with the submission; the allowance criteria by 31 December 2026)

According to the head of government, the actual tax payments of companies with revenue above HUF 180 billion were below 9 per cent and will be around 11 per cent. According to Portfolio, the new tax rate and the basis of the calculation were not disclosed. From raising the corporate tax and the innovation contribution, the government expects HUF 260 billion of additional revenue in total. According to MIAK, this figure is credible if it shows the overlap with the global minimum tax. Under a 2022 directive of the European Union, corporate groups with consolidated revenue of at least EUR 750 million must pay an effective tax of at least 15 per cent in the countries where they operate. Hungary has collected this since 2024 through a domestic top-up tax. Where an affected group already pays top-up tax, the higher corporate tax partly just replaces this top-up. The gross additional revenue can thus be larger than the net. MIAK proposes that the explanatory memorandum of the bill should show how many companies are affected, how many of them fall within the scope of the minimum tax, and how large the additional revenue is after deducting the replaced top-up tax. According to the announcement, the development tax allowance will be “made targeted”. Here MIAK asks that the eligibility conditions be set out by the end of 2026 in a decree, with measurable criteria, and that the allowances granted be public company by company. In Kornai’s framework (see 6.4.3) this is the difference between hard and soft taxation: the allowance should not be the result of a case-by-case bargain. The proposal applies the G3 tax reform and the G20 impact assessment system. On 20 July MIAK already asked for a general tax impact assessment for every tax change. Here it is a concrete interaction not discussed so far.

The three proposals are linked by a common principle: behind the figures presented on Monday there should be a structure that also handles error. The saving should be written into appropriations, there should be a reaction fixed in advance to an error in the growth assumption, and tax revenue should show the amount that can actually be collected. Drucker from the side of decision-making, the IMF from the side of forecasting and Kornai from the side of taxation say the same thing: what is not locked in advance and made measurable turns into a bargain afterwards.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Economy The saving written into chapters and the deviation threshold can reduce the risk premium (spread), i.e. the extra that investors demand for Hungarian government securities. If the chapter cuts are too deep, ministries postpone investments, and this holds back growth.
Society It becomes visible whether the funding of the additional welfare spending comes from a durable saving or a one-off item. In a bad year the deviation threshold may force a rapid adjustment that may also affect social spending.
Public administration The measure-by-measure register makes ministerial financial management transparent. Significant administrative burden; ministries may favour items that are easy to show but of small value.
Corporate sector Showing the minimum-tax overlap gives a realistic revenue expectation, and public allowance conditions reduce case-by-case bargaining. The company-by-company publicity of allowances may trigger a dispute over the protection of business secrets; overly rigid criteria may deter some investments.

The main question for weighing is in proposal 3.2. The in-year deviation threshold strengthens credibility, but it can also make the budget pro-cyclical, i.e. prescribe austerity even during a downturn. MIAK therefore does not propose an automatic cut, but an offsetting proposal within 60 days. This proposal may include a revenue measure, rescheduling, or a justified deviation under the G15 counter-cyclical fiscal stabiliser (public spending in a downturn, saving in an upswing), if the Fiscal Council’s opinion supports it. Proposal 3.1 becomes risky if the chapter cuts worsen the quality of public services. Items affecting services must therefore be marked separately in the register. The risk of proposal 3.3 is low: it asks for a calculation and a rule, not a new tax.

Part V — Measurability and summary

5.1 What is worth following? (suggested KPIs)

MIAK suggests following these performance indicators (KPIs, Key Performance Indicators):

  • Chapter lock-in: whether in the submitted bill at least 90 per cent of the 2,200-billion saving appears as reductions in chapter appropriations. Worth following on the day of submission, expected at the end of October.
  • Mid-year delivery: whether, according to data published by 31 July 2027, at least 50 per cent of the chapter saving has been achieved, and how much of it is durable.
  • Sensitivity table: whether the annex to the bill shows the change in the deficit in the case of a 1 percentage-point deviation in growth, inflation and yields.
  • Net tax revenue: whether the explanatory memorandum of the bill shows the gross revenue effect of the corporate-tax rise and the net effect reduced by the replacement of the top-up tax, together with the number of companies affected.

5.2 Summary

According to the analysts asked by Portfolio and HVG, the 4.9 per cent target is convincing for the markets, so MIAK does not dispute the level of the figure. MIAK asks the Government that, in the bill to be submitted at the end of the month, the 2,200-billion saving be written into appropriations ministry by ministry, and that the 300-billion reserve be available for spending only after mid-year delivery. The bill should come with a sensitivity annex and an in-year deviation threshold, and, for the corporate tax rise, with the net revenue reduced by the minimum tax. Two of MIAK’s foundational values are involved. Accountability, because a saving announced as a single sum cannot be held to account afterwards, whereas one written into appropriations can. And data-drivenness, because the failure of the 2026 assumption of 4.1 per cent showed that a budget’s credibility also depends on how far it reckons with its own error.

Part VI — Reasoning and further sources

6.1 Press framing by spectrum

Liberal-left band. Telex built on analysts’ voices. According to its headline, the budget looks “regime-changing on spending, a cautious reformer on revenue”. The paper highlighted that many important details will not be visible until the bill appears. 444.hu approached from two directions. On the one hand, it reported factually on the adoption of the 2026 amendment. On the other, in an opinion piece it put at the centre the break with “crony capitalism” and a new social bargain in which the wealthy and the circles living off public money get less. HVG’s subscriber analysis wrote about “creative austerity”, while its economics section collected the cautious optimism and risk warnings of analysts speaking to MTI. A relevant Népszava article was not among the processed material.

Public-affairs band. 24.hu concentrated on the figures of the 2026 amendment. Index appeared in two genres: with a concise “in 12 minutes” summary article and with an analyst interview. The latter called the 2,200-billion saving the biggest risk of the budget, and also presented the raising of the middle band of the retail special tax from the perspective of Hungarian-owned shops. ATV carried the assessment of former central bank governor Péter Ákos Bod. He considered the deficit target reassuring, but missed a rethink of utility tariffs. He called the corporate tax rise “timid”, but welcomed the income threshold for welfare benefits.

Economic band. Portfolio gave the most detailed itemised tally. Its analysis described the package as a solution to the “budget trilemma”. It highlighted that the 2.6-point balance improvement would be the largest at EU level, but 1.4 points of it would occur even without measures. In a separate article it presented the assessment of political scientist Gábor Török, who sees left-wing, redistributive elements in the budget, and also named the losers, for example mothers of two children over 40.

Government-critical conservative band. Magyar Nemzet wrote in a frame of “the budget of broken promises”. It highlighted the rising government debt, the large deficit planned for the second half of the year and the cut of a few billion in credit support for small and medium-sized enterprises, building on the arguments of MP Piroska Szalai and Fidesz economic politician André Palóc. Mandiner’s material was not in the processed set.

6.2 Facts and data

Data Value Source
2027 deficit target 4.9% of GDP Péter Magyar, National Assembly, 19 October 2026
2027 growth and inflation assumption 2.4% and 3% respectively Telex, 20 October 2026 (Kossuth Radio)
Deficit target for 2030 2.8% Telex, 20 October 2026
Deficit-increasing decisions c. HUF 2,100 billion Portfolio, 19–20 October 2026; MBH (HVG/MTI)
Deficit-improving package more than HUF 3,000 billion, of which more than two-thirds on the spending side Portfolio, 20 October 2026
Spending saving HUF 700 billion (this year’s, maintained) + HUF 1,500 billion (new) = HUF 2,200 billion Portfolio, 20 October 2026; Index, 20 October 2026
Net balance improvement c. HUF 1,000 billion Portfolio, 19–20 October 2026 (quick calculation)
Balance improvement 2026→2027 2.6 pp, of which 1.4 pp even without measures Portfolio, 20 October 2026
2027 deficit under unchanged policies c. 6.1% Ministry of Finance, July budget inventory (quoted by Telex)
SZJA cut 2.5 million affected; HUF 250 billion revenue loss Portfolio, 19 October 2026
Corporate tax for revenue above HUF 180 billion, actual burden below c. 9% → c. 11% Portfolio, 19 October 2026
Corporate tax + innovation contribution (0.3% → 0.5%) +HUF 260 billion Portfolio; Index, 20 October 2026
Wealth tax c. 15 thousand fortunes affected, HUF 200 billion/year Portfolio, 20 October 2026
Safety reserve HUF 300 billion Index, 20 October 2026
2026 amendment deficit 3.7% → 7.5% (HUF 4,218 → 7,110 billion); growth 4.1% → 2% Telex; 24.hu, 19 October 2026
Vote on the 2026 amendment 134 in favour, 48 against, 1 abstention Telex; 24.hu, 19 October 2026
Government debt 74.6% (end of 2025) → 77.5% (2026) Telex, 19 October 2026
In-year budget amendment since 2015 8th occasion President of the Fiscal Council (444.hu, 19 October 2026)

Two remarks belong to the figures. The 2,200-billion and the 1,500-billion spending cuts are not a contradiction: the first also includes maintaining the 700 billion already achieved this year, the second only the new measures. ATV’s summary wrote about reducing the deficit “from 8.3 per cent” to 4.9 per cent. However, according to the Ministry of Finance, 8.3 per cent is the deficit that would have arisen in 2026 without a change of government. The actual starting value is the 7.5 per cent of the amended law. Péter Ákos Bod spoke of raising the large-company profit tax “from 9 to 11 per cent”, whereas according to Portfolio 11 per cent is the actual burden reduced by allowances, and the new tax rate was not announced. The table follows Portfolio’s wording.

6.3 Policy dimensions

  • Economy (programme points) — writing the saving into appropriations and the spending review (G21), the sensitivity annex and the deviation threshold (G1, G23, G15), the net revenue of the corporate-tax rise and the allowance conditions (G3, G20);
  • Public administration and e-government (programme points) — two-thirds of the saving would come from the state’s own operations; the measure-by-measure register is the practical form of ministerial efficiency measurement (KI8);
  • Social policy (programme points) — doubling the family allowance, raising the minimum pension and the other welfare items are permanent spending, so their funding must come from durable savings; the principle of targeting (SZ1) also applies to marking cuts that affect services;
  • Employment policy (background material) — a high tax wedge, i.e. the share of the employer’s total wage cost going to taxes and contributions, pushes towards undeclared work in low-wage sectors; the SZJA cut in the lower income band may ease this, but its effect is worth measuring on employment data.

6.4 Literature in detail

6.4.1 Peter F. Drucker: The Effective Executive

In the chapter on effective decision-making, Drucker names feedback as the fifth and final element of the decision process. In his view, the mechanism that continuously tests the expectations behind a decision against actual events must be built into the decision from the outset. His reasoning is simple: decisions are made by people, who can err, and even the best decision becomes obsolete at some point. Feedback requires organised information, reports and figures. Drucker adds that this alone is not enough: the decision-maker must go out into the field, because abstract data detached from reality mislead. Assumptions, after all, sooner or later become obsolete. He considers this valid for government policy as well. Translated to the 2027 budget: the 2,200-billion saving is currently a single expectation. Feedback can be built on it only if it breaks down into items, and for each item the baseline amount and the expected result are fixed. The chapter lock-in and mid-year check in proposal 3.1 provide this Druckerian structure.

📖 Source: Peter F. Drucker: The Effective Executive

6.4.2 International Monetary Fund: World Economic Outlook 2023

Box 1.1 of the IMF’s autumn 2023 report examines why medium-term growth forecasts have declined since 2008. The question is whether forecasters have become too pessimistic. Analysing the bias of forecast errors, the authors answer no:

“After the global financial crisis, forecasts were, if anything, biased upwards: realised medium-term growth fell short of what had been forecast.” (MIAK’s own translation)

According to the report, the downward-revised forecasts partly represent a correction of precisely this optimism. Hungarian budget planning shows the same pattern. The 2026 law assumed 4.1 per cent growth, independent analysts expected 1–2 per cent, and the amended law assumes 2 per cent. The 2027 assumption of 2.4 per cent is more cautious, but according to the IMF’s experience the direction of error tends to point downwards. The sensitivity annex and deviation threshold in proposal 3.2 do not dispute the assumption, but fix in advance what should happen if the assumption is not met.

📖 Source: IMF: World Economic Outlook — Navigating Global Divergences (2023)

6.4.3 János Kornai: Economics of Shortage

In Economics of Shortage (A hiány), Kornai describes the hardness of a firm’s budget constraint through several conditions. One of them is taxation. He speaks of hard taxation when the firm cannot influence the shaping of tax rules, the tax is tied to objective, measurable criteria, the firm cannot receive individual exemptions, and the tax is collected on time and unconditionally. Taxation is soft if the firm influences the tax rules, can obtain allowances or deferrals through individual exemptions, and the tax levied is not consistently collected. Kornai stresses that hardness depends not on the level of the rates but on the impersonality of the rules. Although he wrote the work about the socialist economy, he himself indicates that perfect hardness does not fully prevail in market economies either. The 2027 announcement speaks of “making targeted” the large-company development tax allowance. Targeting can take two forms: eligibility tied to public, measurable conditions, or individual negotiation. In Kornai’s framework the first is hard, the second soft taxation. Proposal 3.3 therefore asks for the allowance criteria to be fixed in a decree and for the allowances granted to be public company by company.

📖 Source: Kornai János: A hiány (Economics of Shortage)

6.5 International comparison

Balance improvement in EU comparison. According to Portfolio’s international compilation, in 2027 Hungary would show the largest budget balance improvement as a share of GDP in the European Union, and among OECD members only South Korea would be ahead of it. At the end of his speech, the Prime Minister referred to the examples of France and Romania, where persistent overspending led to political and market tension. The international comparison also shows that large balance improvements are rare, and implementation is the hardest part. This is why the chapter lock-in in proposal 3.1 matters.

Independent macroeconomic forecasting. In several European countries, the macroeconomic assumptions of the budget are prepared or endorsed by an institution independent of the government. In the Netherlands this is done by the Central Planning Bureau (CPB), in the United Kingdom since 2010 by the Office for Budget Responsibility (OBR). The Hungarian Fiscal Council gives opinions, but does not prepare an official baseline forecast. The deviation threshold in proposal 3.2 also works within the existing institutional framework, building on the Council’s public opinion.

Global minimum tax. The 15 per cent minimum tax developed within the OECD and introduced in the European Union by a directive applies to large international corporate groups. The essence of the rule is that if a group’s effective tax burden in a country stays below 15 per cent, the difference must be paid as a top-up tax. This can be collected primarily by the country where the activity takes place. In countries with a low corporate-tax rate, raising the rate for large groups therefore partly just replaces the top-up tax. The actual additional revenue arises from companies outside the scope. The breakdown in proposal 3.3 makes this visible.

Economy

  • G1 — Data-driven budget
  • G3 — Simplification and progressive reform of the tax system
  • G15 — Counter-cyclical fiscal stabiliser
  • G20 — Economic-policy impact assessment system (Drucker audit)
  • G21 — Systematic review of public spending
  • G23 — Government-debt sustainability framework

Public administration and e-government

  • KI8 — Drucker-based efficiency measurement in public administration

Social policy

  • SZ1 — Targeted support

Proposed new programme point: In-year budget deviation threshold — if, according to the quarterly forecast, the deficit exceeds the target by more than 0.3 per cent of GDP, the Government submits an offsetting proposal within 60 days, with a public opinion of the Fiscal Council — for the Economy area.

6.7 List of sources

Press sources (MIAK press monitor, 20 October 2026 — topic 1):

Knowledge-base references (literature):

  • 📖 Peter F. Drucker: The Effective Executive
  • 📖 IMF: World Economic Outlook — Navigating Global Divergences (2023)
  • 📖 Kornai János: A hiány (Economics of Shortage)

MIAK internal materials:

  • MIAK policy area: Economy (programme points; programme point IDs: G1, G3, G15, G20, G21, G23)
  • MIAK policy area: Public administration and e-government (programme points; programme point ID: KI8)
  • MIAK policy area: Social policy (programme points; programme point ID: SZ1)
  • MIAK policy area: Employment policy (background material)
  • MIAK press monitor, 20 October 2026 — topic 1, score: 95/100

Generation metadata