Part I — Situation overview

On 25 August 2026 the European Commission announced that Italy is signing its SAFE loan agreement and will draw down “the large majority” of the envelope of EUR 14.9 billion initially set aside for it; the final amount is expected to come to more than EUR 8 billion. SAFE (Security Action for Europe) is the European Union’s defence-purpose loan programme on preferential terms: member states can finance their own procurements from it, and part of the envelope is expressly set aside for joint procurement by several countries — in that band non-EU Norway may also participate. The background to the announcement is that at the end of July the Commission warned Rome that the delay endangers the organisation of the second lending call, which legally has to close before the end of 2026.

The rules of the second round are being drawn up at present, and from the point of view of the Hungarian decision this is the essential point. According to Euractiv’s report the Hungarian loan agreement has not yet been signed, and the Commission expects residual funds from the Hungarian envelope as well. According to the Commission’s spokesperson no decision has yet been taken on the conditions of the call, and “everything is open” — the key question is whether the second round will also open to countries which did not apply in the first (this would open the way for Ireland, for example). If the call is limited to the countries which have already applied, the envelope distributed in the second round is expected to go to Poland and the Baltic states. The Hungarian decision therefore has not months but weeks.

In the same week the contract that gives the most concrete example of how the programme works was signed in Warsaw. The Polish state-owned Mesko — a member of the PGZ defence industry holding — concluded an agreement worth 8 billion zloty, that is about EUR 1.9 billion, for the supply of shoulder-launched Piorun air defence missile systems for the armed forces of Poland, Norway, Lithuania and Latvia; the delivery contains “several hundred” launchers and “several thousand” missiles, more than 80 per cent of which goes to the Polish armed forces. This is the largest Piorun order ever. According to deputy defence minister Magdalena Sobkowiak-Czarnecka, the Polish Armament Agency is for the first time purchasing not only for its own armed forces but for the allies as well. According to defence minister Władysław Kosiniak-Kamysz, Mesko’s production capacity has grown 2.5-fold since December 2023, and since the launch of SAFE in May the workforce of 3,200 has grown by some 200. Poland is by far the largest SAFE beneficiary, with some EUR 44 billion, and according to the government roughly 90 per cent of the money is spent in domestic industry. The programme has its opponents in Poland too: the right-wing opposition and President Karol Nawrocki argue that the construction gives too much influence to Brussels and imposes a decade-long debt on the country on uncertain terms.

MIAK’s reading: the Hungarian question at present is not whether “SAFE is good”, but whether we take the decision at all before the rules of the second round are closed. Postponement is not a neutral option in this structure — if a residue arises from the Hungarian envelope and the second call narrows to the countries which have already applied, the funding goes elsewhere. Rejecting the loan may be a legitimate decision; failing to take the decision is not.

Part II — Foundations in the literature

Three volumes help to ensure that the present news can be read as more than a mere procurement item. In his work World Order Henry Kissinger (American diplomat and thinker on foreign policy, former Secretary of State) sharply separates alliance from collective security: an alliance is an agreement built on concrete facts and expectations, creating a precise obligation for defined situations, whereas collective security is a legal construct which defines no concrete obligation at all. The Polish joint procurement contract falls unambiguously into the first category in this distinction — and that is what makes the news a policy event. The volume 23 Things They Don’t Tell You About Capitalism by Ha-Joon Chang (a development economist born in South Korea, teaching in Cambridge and London) examines the question of the conditions under which a government can successfully develop a sector: on his answer it is not intention but the informational channel maintained with the business sector and the regular reporting obligation of the supported firms that decides. And The Innovator’s Dilemma by Clayton M. Christensen (American business economist, originator of the theory of disruptive innovation) explains why it was precisely a shoulder-launched, relatively cheap system that became the export hit of the Polish defence industry: simpler, cheaper, more reliable products regularly override expensive, integrated platforms if they are good enough for the given task. The detailed treatment of the literature — author by author, with quotations — can be found in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures. None of them takes a position for or against taking out the loan — assuming debt is a competence of the government and of Parliament. The proposals are aimed at ensuring that the decision is taken before the finalisation of the rules of the second round, in a documented and quantified way.

3.1 A public, quantified SAFE decision document (before the finalisation of the rules of the second round, at the latest by October 2026)

The government should publish a public decision document of not more than twenty pages answering three figures. First: how large the actual interest advantage of the SAFE loan is relative to market financing, over the maturity, expressed in forints — this figure is not known today, even though the sense of the whole programme depends on it. Second: what percentage of the envelope that can be drawn down may be spent at a domestic supplier under the programme rules in force, and alongside what industrial capacity — the Polish reference value is 90 per cent, the realistically attainable Hungarian ratio may differ from this, but it has to be named. Third: how large the debt path effect of the loan is over the next ten years, according to the methodology of the G23 public debt sustainability framework. At the end of the document there should be an unambiguous recommendation, and — if the result is the rejection of participation — its reasoning as well. This is the direct application of the HV2 defence expenditure transparency programme point to a concrete decision opening now.

3.2 A Hungarian consortium application to the joint procurement band (within 60 days of the opening of the second call)

The joint procurement band is that element of SAFE in which it is not the size of the envelope but the industrial position that decides: according to the Polish example, a state-owned manufacturer can become the leader of a consortium supplying the armed forces of three allies. Hungarian defence industry — with the Rheinmetall plant in Zalaegerszeg and the domestic supplier base — can realistically enter this band not as a consortium leader but as a supplier, and for that the Ministry of Defence and the domestic manufacturers have to be ready by the time the call opens: there should be an up-to-date, English-language capacity and qualification register, and a designated contact person towards the procurement agencies of the partner countries. This is the most concrete implementation window of the HV4 programme point that is opening now: the programme point sets a target of at least 20 per cent of Hungarian defence procurement being conducted in joint EU procurement by 2028, and the integration of more than 30 domestic small and medium-sized defence enterprises into the European supply chain.

3.3 An annual domestic supplier ratio report on defence procurements (from the first quarter of 2027)

If Hungarian participation takes place, an annual, public report should be prepared on the domestic supplier ratio of subsidised or loan-financed defence procurements: how large a sum went to a domestic manufacturer, how large to a foreign one, and how many Hungarian enterprises entered as new suppliers. On Chang’s argument (see 6.4.2) it is precisely this regular reporting obligation that distinguishes successful industrial policy from unsuccessful: this is how the government obtains the information without which it cannot choose well. The report would be the shared measuring instrument of the HV3 defence innovation programme and of the G9 strategic industrial policy programme point — and at the same time the only data from which it can subsequently be judged whether taking out the loan was worth it.

The three proposals are linked by a single principle: defence borrowing is defensible if it turns into industrial investment, and not if it turns into import financing. The Polish case shows this difference clearly — there, arguments are made with figures both for and against the programme, and it is precisely this that makes the debate public. In Hungary there are at present no public figures on either side, and this absence is in itself a decision.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Economy The public calculation of the interest advantage and of the debt path makes it possible to judge the decision on economic rather than political grounds; in the event of an industrial supplier position, a lasting order book SAFE is preferential, but it is a loan: repayment is a decade-long obligation, and if the funding goes on import procurement, the domestic economic return is minimal
Society The transparency of defence expenditure reduces the mistrust surrounding the programme; supplier expansion means qualified industrial jobs Raising defence expenditure narrows the room for manoeuvre in other sectors; the Polish debate shows that this is a legitimate and lasting political dividing line
Public administration The decision document and the capacity register create institutional knowledge that can be reused in later procurement cycles Alongside a deadline measured in weeks, quality preparation may run into a shortage of capacity; a poor-quality, rushed decision is worse than a negative one

The main trade-off is between time and the quality of the decision. The legal deadline for the rules of the second round is the end of 2026, and the framework rules are being written now: whoever does not speak up now will not have the rule take account of them later. At the same time, a decision assuming a decade-long debt obligation must not be taken in a few weeks without an impact analysis. The road between the two is that the decision document should fix not the fact of taking out the loan but its conditions: alongside what domestic supplier ratio and what interest advantage it is worth it, and what happens if these are not met. This structure makes a fast yet documented decision possible. The proposal tips to the risk side if the decision document becomes the subsequent justification of participation: the figures will then not be an analysis but a rationalisation, and the public presentation of counter-arguments prescribed by the G19 programme point will be missing.

Part V — Measurability and summary

5.1 What is worth following? (proposed KPIs)

MIAK proposes the following performance indicators (KPIs) for monitoring. These are proposals, not government commitments.

  • Decision turnaround: how many days before the finalisation of the rules of the second round the public Hungarian decision document appeared — proposed target: at least 30 days earlier.
  • Domestic supplier ratio: what percentage of the defence procurement financed from SAFE (or, in the event of its rejection, from alternative financing) went to a domestic manufacturer — the HV4 programme point sets a target of 20 per cent of procurement in joint EU procurement by 2028; for the domestic supplier ratio the Polish 90 per cent is an upper reference value, not a target value.
  • Consortium presence: how many Hungarian enterprises appear in at least one SAFE joint procurement contract as a supplier or subcontractor — proposed monitoring horizon 24 months.
  • Expenditure transparency: whether an annual, public defence procurement supplier report appears — proposed target: yes, from the first quarter of 2027.

5.2 Summary

MIAK’s key message: the decision on Hungarian SAFE participation has to be written down, with figures, still before the finalisation of the rules of the second round. We ask the government to publish, by October 2026 at the latest, a public decision document on the actual interest advantage of the loan, on the share that can be spent at a domestic supplier and on the debt path effect; and to prepare the application of Hungarian industry for the joint procurement band. If the result of the analysis is that participation is not worth it, that is a legitimate decision — but then the document should also contain an alternative financing plan for building domestic defence industrial capacity. We ask the public not to look at the question in a “Brussels money versus national sovereignty” frame: SAFE is a preferential loan whose assessment is a matter of interest calculation and industrial capacity analysis.

Of MIAK’s foundational values two are in play here. Data-drivenness, because judging defence borrowing depends solely on figures — the interest advantage, the domestic supplier ratio and the debt path effect can all be calculated, and without these the debate necessarily slides onto a symbolic plane. And transparency, because defence expenditure is the area where secrecy is most often warranted — and precisely for that reason the most easily over-extended. Protecting concrete capability data is legitimate; protecting the interest conditions of a financing construction is not. Separating the two is the point of the HV2 programme point.

Part VI — Justifications and further sources

6.1 The framing of the press, spectrum by spectrum

The band of international specialist portals carried two complementary narratives. Euractiv’s article “Italy to access more than €8bn in SAFE defence loans” worked in an institutional and procedural frame: it made the improvement in the relationship between the Commission and Rome the axis of the story, and linked to it the organisational pressure of the second lending call. It is in this article that Hungary also appears — not as a topic in its own right, but in the list of unsigned contracts and expected residual funds. The paper’s “The Brief” column carried the same news in a single sentence, in summary form. This framing is the most important one for the Hungarian reader: it presents the situation not as a conflict but as a timetable, and from this follows the time pressure on the decision.

Notes from Poland’s piece “Poland’s Mesko signs €1.9bn deal to supply Piorun air-defence systems to four NATO countries”, by contrast, framed the event as an industrial policy success story: the backbone of the article is the 2.5-fold expansion of production capacity, the growth in headcount and the 33 per cent rise in 2025 revenue, while the ministerial quotations emphasise the “European leader” position. At the same time the paper — and this shows the quality of the treatment — carries the counter-arguments in a separate paragraph as well: the criticism of the right-wing opposition and of President Karol Nawrocki, according to which the programme gives too much influence to Brussels and imposes a decade-long debt on the country. This duality does not appear at all in Hungarian public discourse at present: there are no public Hungarian figures on either the opportunity or the risk side.

Deutsche Welle’s material “German army helps Poland protect itself against Russia” supplies the third level: the physical reality of the Polish defence effort — the construction of the roughly 800-kilometre East Shield, the defence budget raised to 200 billion zloty for 2026, that is to nearly 5 per cent of gross domestic product (GDP), and the participation of German engineers since June. According to the monitor, the Hungarian press did not carry this topic in a treatment of its own on that day; in domestic public life the second SAFE round is at present not a topic. This absence is the most essential finding at the level of press framing: a decision with a deadline measurable in weeks and creating a decade-long obligation is under way without any band of the domestic spectrum putting it on the agenda.

6.2 Facts and data

Fact Value Source
Italian SAFE envelope EUR 14.9 billion, of which more than EUR 8 billion expected to be drawn down European Commission, 25 August 2026
Legal deadline of the second lending call end of 2026 Warning of the European Commission, end of July 2026
Status of the Hungarian loan agreement unsigned, residual funds expected Euractiv, 25 August 2026
Mesko–Piorun contract 8 billion zloty (approx. EUR 1.9 billion) Notes from Poland, signature of 24 August 2026
Buyers Poland, Norway, Lithuania, Latvia (more than 80% of the quantity is Polish) Notes from Poland
Mesko production capacity 2.5-fold expansion since December 2023 Defence minister Kosiniak-Kamysz
Mesko headcount approx. 3,200 + approx. 200 expansion since the launch of SAFE in May 2026 Defence minister Kosiniak-Kamysz
Mesko revenue (2025) 2.3 billion zloty, +33% year on year Notes from Poland
Polish SAFE envelope approx. EUR 44 billion, the largest beneficiary Notes from Poland
Domestic use in Poland approx. 90% Polish government statement
Polish defence budget (2026) 200 billion zloty (approx. EUR 46 billion), close to 5% of GDP Deutsche Welle
Length of the East Shield approx. 800 kilometres (planned) Deutsche Welle

6.3 Policy dimensions

  • Defence (programme points) — accession to EU joint procurement initiatives, integration of the domestic supply chain and transparency of defence expenditure (programme point ID: HV4, HV2, HV3);
  • Defence (background material) — the doctrine of independent defensive capability: allied joint procurement and national capacity building are not alternatives but presuppose each other (programme point ID: HV9);
  • Economy (programme points) — strategic industrial policy and the quantification of the debt path effect (programme point ID: G9, G23, G10);
  • Foreign policy (programme points) — issue-based coalition building: the joint procurement band creates partnership in a concrete, contractual form (programme point ID: KP17).

6.4 Literature in detail

6.4.1 Henry Kissinger: World Order

One of the most precise conceptual distinctions in Kissinger’s volume runs between alliance and collective security. On his argument an alliance is an agreement built on concrete facts or expectations: it creates a precise obligation to act in defined situations, it derives from interests recognised as common, and the more parallel those interests are, the stronger the cohesion of the alliance. Collective security, by contrast, is a legal construct addressed to no specific situation.

“An alliance comes about as an agreement on specific facts or expectations. It creates a formal obligation to act in a precise way in defined contingencies. […] Collective security, by contrast, is a legal construct addressed to no specific contingency.” (Henry Kissinger: World Order)

This distinction gives the Polish joint procurement contract its real weight. When the Polish Armament Agency purchases not only for its own armed forces but for Norway, Lithuania and Latvia as well, then in Kissinger’s sense what is born is not a declaration of solidarity but an alliance obligation: a common production line, a common logistics chain, a common spare parts supply. The dependence thus created is two-way and contractual — and precisely for that reason lasting. From the Hungarian point of view it follows that participation in the joint procurement band is not merely a procurement question: this is the form in which the alliance relationship acquires concrete, measurable content. The KP17 issue-based coalition building programme point is built on exactly this mechanism.

📖 Source: Henry Kissinger: World Order

6.4.2 Ha-Joon Chang: 23 Things They Don’t Tell You About Capitalism

One chapter of Chang’s volume refutes the widely held proposition that governments are incapable of taking well-founded sectoral development decisions. First he presents the counter-argument in its strong form — government decision-makers are motivated by power, not by profit, and it is not they who have to bear the financial consequences of their decisions — and then he goes through those cases in which active industrial policy nevertheless worked: alongside Korea, Taiwan and Singapore also France, Finland, Norway and Austria, and further the federal research and development support of the United States, to which the computer, semiconductor, aircraft and biotechnology industries all owe their development.

The most important finding of the book — and the one directly applicable to the present topic — is, however, not that industrial policy can work, but what makes it work.

“The evidence shows that those governments had a more successful policy of differentiated support which possessed more effective channels for the exchange of information with the business sector.” (Ha-Joon Chang: 23 Things They Don’t Tell You About Capitalism, quoted from the Hungarian edition)

Chang names three concrete instruments for obtaining information: the state-owned enterprise, the obligation of firms in supported sectors to report regularly, and networks between officials and business leaders — and on the last of these he separately warns that the exclusivity of the channel may lead to cliques or to corruption.

The Polish case shows all three elements: Mesko is a member of a state holding, the government argues for the programme with quantified production and headcount data, and the procurement decisions are coordinated by a dedicated agency. The condition for taking this over in Hungary is the same, plus the handling of the risk indicated by Chang: if the selection of the domestic supplier base takes place through an informal network, the programme achieves precisely the effect it is trying to avoid. Hence proposal 3.3 — the annual, public supplier ratio report is at once an information channel and a reducer of corruption risk.

📖 Source: Ha-Joon Chang: 23 Things They Don’t Tell You About Capitalism (Hungarian edition: 23 dolog, amit nem mondtak el a kapitalizmusról)

6.4.3 Clayton M. Christensen: The Innovator’s Dilemma

The central observation of Christensen’s book is that market leaders do not lose position because they manage themselves badly, but because the logic of good management follows the needs of existing, well-paying customers — while the real threat arrives from a simpler, cheaper technology that at first performs worse in the main market.

“Disruptive technologies […] are typically simpler, cheaper, and more reliable and convenient than established products.” (Clayton M. Christensen: The Innovator’s Dilemma)

To this the volume adds the decisive condition: a disruptive technology breaks into the main market when it reaches the “good enough” level for the given task — in the example of hydraulic excavators, when their bucket capacity reached the level expected by the market, even if the old technology could bear more.

Shoulder-launched air defence systems are following exactly this path. They do not compete with integrated, layered air defence systems in high-altitude, long-range tasks — but against low-flying aircraft, helicopters and drones they are good enough, while being orders of magnitude cheaper, quick to mass-produce, and lower in training requirements. In Christensen’s frame this explains how a medium-sized Polish state manufacturer became the supplier of three allies within three years. From the Hungarian point of view it follows that the entry point into the defence industry is not necessarily at the level of the large platforms: cheap, mass-producible, quickly scalable systems and their components are the segment in which a smaller industrial base can also be competitive. The HV3 defence innovation programme point follows this logic of entry.

📖 Source: Clayton M. Christensen: The Innovator’s Dilemma

6.5 International comparison

There are two established European models for joint defence procurement, and SAFE is essentially a blend of the two. One is the NATO Support and Procurement Agency (NSPA), which for decades has carried out joint procurement on member state mandate from ammunition to fuel; its advantage is economies of scale, its limitation that it does not finance, only coordinates. The other is the project-based cooperation framework of the European Defence Agency (EDA), which aligns capability development, but without funding of its own. The novelty of SAFE is that it attaches a preferential loan to the coordination as well, and that in the joint procurement band it admits a non-EU member — Norway. According to the data of the HV4 programme point, consolidation within Europe may produce a unit cost reduction of as much as 50–90 per cent for identical weapon systems, and the same programme point warns of the reverse risk: if procurement at EU level prefers the larger German and French manufacturers, Hungarian industry may be squeezed out. In the present situation this risk has taken concrete form — if the second call narrows to the countries which have already applied, the residue goes to Poland and the Baltic states. The lesson of the Polish model is not that it is easy with an envelope of EUR 44 billion: it is that industrial preparation — capacity expansion, qualifications and allied relationships — preceded the signing of the contract by years.

Defence

  • HV4 — EU defence industrial base and joint procurement
  • HV2 — Transparency of defence expenditure
  • HV3 — Defence innovation programme
  • HV9 — Independent defensive capability, the “one’s own arms” doctrine

Economy

  • G9 — Strategic industrial policy
  • G23 — Public debt sustainability framework
  • G10 — State development bank
  • G19 — Radical transparency in economic decision-making

Foreign policy

  • KP17 — Issue-based coalition building in the EU

Proposed new programme point: Defence procurement supplier ratio report — for the Defence area: an annual, public statement on the domestic and foreign supplier breakdown of defence procurements and on the number of newly entering Hungarian enterprises.

6.7 List of sources

Press sources (MIAK foreign press monitor, 26 August 2026 — topic 2):

Knowledge base references (literature):

  • 📖 Henry Kissinger: World Order
  • 📖 Ha-Joon Chang: 23 Things They Don’t Tell You About Capitalism
  • 📖 Clayton M. Christensen: The Innovator’s Dilemma

MIAK internal materials:

  • MIAK policy area: Defence (programme points; programme point ID: HV4, HV2, HV3, HV9)
  • MIAK policy area: Economy (programme points; programme point ID: G9, G23)
  • MIAK policy area: Foreign policy (programme points; programme point ID: KP17)
  • MIAK foreign press monitor, 26 August 2026 — topic 2, score: 89/100

Supplementary public data sources:

  • European Commission — SAFE member state allocation table and signature status
  • European Defence Agency (EDA) — joint procurement project list
  • SIPRI — Arms Industry Database
  • NATO — annual defence expenditure data release

Generation metadata