Ukraine is centralising its war machine just as EU accession demands the opposite. It faces an impossible choice.
A sweeping reorganisation of Ukraine’s foreign and defence apparatus in the space of a few August days has faced scrutiny almost exclusively through a political lens. Who is up, who is down, what did Mykhailo Fedorov do wrong (or right), what does Volodymyr Zelensky want? What does it signal to Washington, Brussels, and Moscow?
Those are legitimate questions. But they are not the only ones. Viewed through the less partisan lens of system management theory, the reshuffle raises a different set of problems: not about intentions, but about architecture. In some ways, it is more consequential than its politics. Is Ukraine building a system that can actually do what it needs to do?
What happened in Kyiv
Can Ukraine win a war and join the Union simultaneously? Or is it optimising for one goal in ways that structurally undermine the other? Putting it simply: when the dust settled, the war may have left Ukraine with some features that sit uneasily alongside its European trajectory.
The reshuffle was the most extensive reorganisation of Ukraine’s foreign and security establishment since the full-scale invasion began. The foreign ministry, the defence ministry, and several of the coordinating bodies that link them to the presidential office underwent a hasty restructuralisation. A clear directional logic ruled the change: tighter central control, shorter chains of command, and more direct accountability to the top.
You might be interested
Coordination of public administration reform, previously a domain of several ministries, moved under the deputy prime minister for European and Euro-Atlantic integration. Several experienced negotiating teams do not exist any more, or have taken a very different shape. The signal, whether intentional or not, was unmistakable: Kyiv has had enough of the loose, pluralistic, sometimes chaotic reform ecosystem that characterised its pre-war and early-war governance.
It is moving toward something harder, faster, and more unified. Politically, this has been interpreted as Mr Zelenskyy consolidating power ahead of a difficult diplomatic season. From a systems perspective, the interpretation is both more interesting and more ambiguous.
The centralisation question
The foundational question in system management is not whether to centralise or decentralise, but how much of each, and where. The most rigorous treatment of this problem in the academic literature is the work modelled in Factors that affect the optimal amount of central control in complex systems by Daniel Solow and Joseph G. Szmerekovsky, published in the US Naval Research Logistics journal. It argues that the optimal level of central control in any complex system depends on a combination of factors.
Those factors are: the volatility of the environment, the interdependence of subsystems, the cost of coordination failures, and the capacity of central nodes to process information accurately and quickly. No single architecture is universally superior. The right answer is always conditional. “The optimal amount of central control depends on the system’s environment and the interactions among subsystems,” the authors say.
Applied to Ukraine in August 2026, the conditional factors are stark. The environment is maximally volatile: active war, shifting alliance dynamics, an unpredictable American posture, and a Russian adversary that exploits any coordination gap it can find. The subsystems, military, diplomatic, intelligence, and reform, are deeply interdependent: what the foreign ministry negotiates affects what the defence ministry can procure.
The Brandts test
That in turn affects what the reform agenda can credibly promise to Brussels. The cost of coordination failures is existential. And the central node, the presidential office, has demonstrated over four years of full-scale war a genuine capacity to process information and make decisions quickly. On all four criteria, the model points in the same direction: higher central control is justified. This is not a political judgement; it is what the variables predict.
Experimental findings reinforce the point from a different angle. “Contrary to the theory, overall performance is higher under centralization,” says the Centralized vs. Decentralized Management: An Experimental Study by Jordi Brandts and David J. Cooper. “Mis-coordination is common under decentralization and divisions fail to use their information when they do coordinate,” the authors found. They are careful to note that this is not universal: in environments where local knowledge matters more than coordination, decentralisation wins.
But in Ukraine’s current situation, where the premium is on speed, coherence, and the ability to present a unified face to multiple external interlocutors simultaneously, the Brandts findings support the reshuffle’s internal logic.
The limits of the centre
So far, the case for what Kyiv has done looks strong. The complication arrives with the third body of work: the core-periphery model developed in Centralised vs. Decentralised Management of Public Information Systems. Its central argument is that neither pure centralisation nor pure decentralisation is optimal in complex, multi-stakeholder environments.
The optimal amount of central control depends on the system’s environment and the interactions among subsystems.
— Factors that affect the optimal amount of central control in complex systems, Naval Research Logistics
Pure centralisation delivers coherence but creates bottlenecks, suppresses local knowledge, and, critically, makes the whole system fragile to failures at the centre. Pure decentralisation avoids those pathologies but produces waste, duplication, and coordination failures of its own. The most effective architecture, the authors argue, is a hybrid. “A successful core-periphery approach can simultaneously provide the control necessary to share key resources (…) and the freedom necessary to meet user needs,” the University of Manchester team wrote.
This is where the Ukraine reshuffle becomes genuinely problematic from a systems perspective, and where the analysis diverges from the political cheerleading on both sides. The question is not so much whether Kyiv needed more central control: it probably did. The question is whether it has retained enough functional periphery.
The cost of thinning the periphery
The reform ecosystem that Ukraine built between 2014 and 2022, the anti-corruption institutions, the civil society networks, the specialised negotiating teams, the technocratic clusters inside ministries, was precisely the kind of periphery the core-periphery model prescribes. It carried knowledge, credibility, and implementation capacity that no central node can replicate. If the reshuffle has thinned that periphery in the name of command coherence, it has traded a long-term systemic asset for a short-term operational gain.
The papers make a related point in more formal terms: central control becomes counterproductive when the central node’s information-processing capacity is exceeded by the complexity of the environment. Ukraine’s environment, managing a war, an accession process, a coalition of dozens of partner states, and a domestic reform agenda simultaneously, is about as complex as any state has faced in recent memory.
The risk is not that centralisation is wrong in principle. The risk is that it is being applied at a scale that exceeds what any central node can reliably manage.
How this plays out with EU institutions
This is where the system management analysis produces its most counterintuitive result, because the EU is itself a complex system, and one with a very specific architecture that is almost the inverse of what Kyiv is building.
The EU is, structurally, a core-periphery system of extraordinary elaboration. The Commission is the closest thing to a centre, but it is a centre with limited direct power: it proposes, monitors, and enforces, but it does not command. The Council is a forum of 27 national peripheries, each with veto rights in sensitive areas. The Parliament is a third node with its own mandate and its own information flows.
Directorate-General for Neighbourhood and Enlargement Negotiations (DG NEAR) and the enlargement machinery sit inside the Commission but operate with considerable technical autonomy from political leadership. The rule-of-law and financial-control ecosystem is deliberately insulated from political interference. The whole architecture is designed to prevent any single actor from dominating the system, which is precisely what makes it change-resistant, slow, and, from Kyiv’s perspective, maddening.
A structural collision
The collision between Ukraine’s newly centralised architecture and the EU’s deliberately distributed one is not just political. It is structural. A centralised interlocutor finds it genuinely difficult to interface with a distributed system, because there is no single point of entry. DG NEAR will respond to capability-building and administrative simplification packaged in the language of the accession benchmarks.
The Commission’s rule-of-law directorate will apply its own criteria, independently of what DG NEAR thinks. The Council’s enlargement working group will ask different questions from the foreign-affairs machinery. The Parliament’s Ukraine rapporteurs will have different priorities again. Ukraine’s new, tighter command structure may be excellent at producing a unified message, but the EU system is not designed to receive a unified message. It is designed to interrogate it from multiple angles simultaneously.
Mis-coordination is common under decentralization and divisions fail to use their information when they do coordinate.
— Centralized vs. Decentralized Management: An Experimental Study, Jordi Brandts and David J. Cooper
The granular picture is this. The Commission is the most likely to accept Ukraine’s centralisation if it demonstrably improves delivery on the accession benchmarks: merit-based civil service, anti-corruption institutions, financial control, the 2026–2030 public-administration roadmap. It will not accept it if it looks like consolidation without institutional safeguards. DG NEAR’s enlargement specialists will be watching implementation data, not press releases.
The EU and the periphery question
The Council is structurally harder, because unanimity rules give several capitals, not only Budapest, the ability to slow or block progress if they conclude that Kyiv’s centralisation is politicising appointments or weakening independent institutions. The Parliament will be the most rhetorically supportive of Ukraine’s security role and the most likely to punish anything that reads as reform fatigue or elite capture.
In systems terms, Ukraine’s optimal interface strategy with the EU would be precisely the core-periphery model it appears to be moving away from: a strong central coordinator to manage the overall accession narrative, combined with a resilient, technically capable periphery that can engage each EU node in its own language. The reshuffle may have strengthened the coordinator. The question is what it has done to the periphery.
The conclusions that follow from this analysis, alas, are uncomfortable for both sides. Actually, they are equally unfomfortable for the systems management theory itself. It mostly fails to consider dynamic scoring, i.e., the notion that the actors’ behaviour is likely to change over time, even if technically their position remains the same. The idea is perfectly plausible that a leader should concentrate power in his hands for legitimate reasons, yet later simply holds on to the position for the power per se. Just ask those who have lived under or Recep Tayyip Erdoğan in Turkey or Saddam Hussein in Iraq.
What both systems need to change
Ukraine needs to resist the temptation to treat centralisation as a terminal state rather than a wartime instrument. The Factors model is explicit: optimal central control is a variable, not a constant. As the environment changes, if and when the war moves toward negotiation, or accession talks enter their most technically demanding phase, the system will need to redistribute authority back toward the periphery.
Institutions that have been hollowed out are hard to rebuild. People who have been sidelined do not always come back. The reformist networks that gave Ukraine its unusual credibility in Brussels between 2014 and 2022 were built over years. They cannot be reconstituted by decree.
The EU, for its part, needs to confront the dysfunction of its own architecture more honestly than it usually does. A system designed to prevent domination is also, by construction, a system that is slow to respond to genuine emergencies. The accession process as currently designed is institution-heavy, rule-bound, and calibrated for peacetime candidates with the luxury of gradual alignment. Ukraine is not that.
The central conclusion
The EU’s own Factors problem is that its central nodes, the Commission, the Council presidency, the high representative, lack the authority to make the system move faster even when the political will exists. Several member states have proposed reforms to the enlargement methodology that would allow more differentiated, faster-track integration in areas where Ukraine is already aligned. Those proposals have stalled in exactly the kind of distributed veto environment the core-periphery model predicts.
The Kyiv reshuffle, assessed as a system management decision rather than a political one, is probably the right call for winning the war. At the same time, it may be the wrong architecture for joining the union. Both judgements may be simultaneously correct.
A successful core-periphery approach can simultaneously provide the control necessary to share key resources (…) and the freedom necessary to meet user needs.
— Centralised vs. Decentralised Management of Public Information Systems, University of Manchester
The war requires what the Brandts findings support: centralised management of interdependent tasks in a high-cost, high-volatility environment. The accession process requires what the core-periphery model prescribes: a resilient, distributed institutional periphery capable of engaging a complex, multi-node external system on its own terms. These are not easily reconcilable requirements. A state that optimises for one is likely to underperform on the other.
The deadly paradox
The deepest irony may be this: the qualities that have made Ukraine’s resistance so remarkable, adaptability, institutional creativity, the energy of a reform ecosystem that was genuinely bottom-up, are precisely the qualities that a centralising reshuffle puts at risk. And the qualities that Brussels most rewards in an accession candidate, institutional resilience, independent oversight, decentralised implementation capacity, are precisely what a wartime command structure tends to erode.
The potential tragedy is that Kyiv does not have the choice between one or the other. To survive, it needs Europe’s good will just as much as it needs its steely (if messy) resolve vis-à-vis Moscow; both at the same time.
Ukraine may be building exactly the machine it needs to survive the next two years. The issue is whether that machine can then be, at once, the slick princess Europe requires and the dirty dragon the war needs. It may be an order taller than any country (let alone the European Union itself) could ever fulfil. And for sure it is a question that no reshuffle, however well-designed, can answer in advance.
