Nemzeti Kutatási, Fejlesztési és Innovációs Hivatal

Indoklás

This is the largest line in the chapter — over twice the size of the entire Hivatal — and it is analytically the most straightforward. It pays Hungary's treaty-based subscriptions to international research infrastructures: CERN (member since 1992), the European Molecular Biology Laboratory (since 2017), ITER (since 2007), the European Spallation Source (since 2014), the European Synchrotron Radiation Facility (since 2017), European XFEL (since 2009), ELI ERIC (since 2010), and a set of smaller ERIC consortia and cooperation programmes. The framework's three questions resolve cleanly here, and the resolution is Keep. **Voluntariness:** the fee buys Hungarian researchers, firms and universities access to capital-intensive scientific infrastructure — a particle accelerator, a fusion reactor, a synchrotron — that no national budget could replicate and that exists only because dozens of states pool the cost. The membership fee is not a transfer to a domestic constituency; it is the price of access to a shared facility, set by international treaty, and the alternative is not "voluntary financing of the same thing at lower cost" but exclusion from the infrastructure entirely. **Calculation:** the fee schedule is not a price the Hungarian state sets through a domestic political process where the calculation problem would bite — it is a treaty parameter, typically scaled to member-state GNI, determined by the international organisation. Hungary either pays the scheduled fee and retains access or withdraws; there is no domestic discretion to misprice. **Public-choice exposure:** the membership does not concentrate a rent on an organised domestic lobby in the way a discretionary grant pool does. The beneficiary is genuinely diffuse — every Hungarian physicist, materials scientist or molecular biologist who uses beam time — and the cost, at 17,536.7 millió Ft spread across the working tax base, is genuinely small per head. The full envelope works out to a few thousand forints per year for a full-time taxpayer (illustratively, 17,536.7 millió Ft over roughly 2.2 million full-time-equivalent taxpayers is on the order of 8,000 Ft each), against access to research infrastructure with a replacement cost no national budget could carry. Two honest caveats keep this from being a reflexive Keep. First, the line is governed by treaty: each membership has its own accession instrument and, generally, its own withdrawal provision, so a future government that judged a specific membership poor value could exit that organisation on the treaty's notice terms. Keep here is a judgement that the *current* portfolio is good value, not a claim that the line is permanently fixed. Second, the fees are denominated in EUR and CHF, so HUF volatility against those currencies in recent years has raised the forint cost of the portfolio independently of any change in the underlying fee structures — a budgeting exposure worth naming, though not a reason to reclassify. The line is a Keep: it finances access to a class of capital-intensive research infrastructure that is, almost by construction, beyond the reach of any single mid-sized economy acting alone.

Indoklás

This is the payroll of the office that coordinates national RDI strategy and administers the competitive-grant system. The honest classification question is not whether *research* should be state-financed — it is whether the *coordinating administrator* is a rights-protection function, a constitutional precondition, or a protective response to irreversible harm. It is none of these. It is a discretionary policy office. That argues against Keep. But the calculation logic of the framework also cuts against treating the office as pure rent: a body that runs a grant competition, administers international-treaty memberships, and issues the R&D certifications that determine corporate-tax treatment is performing real administrative work with real counterparty reliance, and abolishing the office without first reforming the grant system it administers would strand those functions mid-flight. The defensible position is a nominal freeze: hold payroll flat, let real-terms erosion of roughly 20-25% over a decade at 2.5% inflation discipline the headcount, and revisit the office's scale as part of a broader reform of how — and whether — the state allocates competitive research subsidy. The Hivatal employs 234 authorised positions (27 leadership, 190 non-managerial, 17 vacant as of early 2026).

Indoklás

This line deserves separate scrutiny precisely because of what its label does not say. "Egyéb működési célú kiadások" — "other operating-purpose expenditures" — is 2,000.1 millió Ft, a sum larger than the office's entire operating budget (1,733.4 millió Ft) and roughly two-thirds the size of its payroll, parked under a heading that describes nothing. A line that funds a defined, nameable activity gets named. A line that is 26% of the office's own-account spending and is called "other" is, on the evidence of its own label, a discretionary pool — the kind of appropriation that exists to be allocated at the holder's judgement during the year rather than against a costed plan fixed in advance. Where the state holds a pool of money whose use is not specified at the point of appropriation, the use is decided later, by officeholders, without a price signal telling them what the marginal forint is worth in one direction versus another; the allocation is subjective, and a subjective allocation of involuntarily-collected revenue is exactly the pattern the framework treats as rent rather than function. The burden of proof runs the other way: a 2,000.1 millió Ft "other" line is presumptively a phase-out unless the office can itemise it into specific, costed, defensible activities — at which point those activities should be appropriated under their own names and classified individually. Until that itemisation exists, the honest treatment is a short phase-out, giving the office three budget cycles either to fold genuinely necessary spending into the named operating line or to let the discretionary residue lapse.

Indoklás

Materials, utilities, IT, services and the recurring operating cost of running the office. It tracks the office's scale; with the office held to a nominal freeze, its operating budget carries the same treatment. The administrative cost of separately litigating an operating-budget cut would exceed the saving, and the bounded nature of the function does not justify expansion.

Indoklás

Capital spending — equipment, IT infrastructure, fixed assets — for the office. Notably, this 573.3 millió Ft capital line is matched almost exactly by 495.5 millió Ft of capital revenue, so the office's net call on general revenue for capital is only 77.8 millió Ft. This is a small, self-limiting, largely self-financed line. There is no case for a discretionary cut whose administrative cost would exceed its saving, and no case for expansion. Nominal freeze.

Indoklás

The employer-side payroll levy on the personnel line above. It moves mechanically with that line and carries the same classification. Worth naming what this number is, because it is the same wedge that runs through every chapter of the budget: the state, as employer, pays SzocHo on top of gross salary, and that 396.2 millió Ft is a tax the state levies on its own payroll and then collects from itself — a closed loop here, but the identical mechanism that on a private payroll suppresses the worker's take-home before it is ever paid. For a private employer, roughly 41 Ft of every 100 Ft of total employer cost reaches the state before the worker spends anything (income tax + employee social security contributions + employer SzocHo, combined share of total labour cost at the average wage); ÁFA on what remains takes a further 11-12 Ft of the original 100; excise on fuel and energy more on those categories. The cumulative effective state take on a typical working household's full employer compensation sits in the 52-55% range. This line is one small visible instance of a wedge that the budget mostly leaves invisible.