–Gazmend Demolli, Independent researcher
A republic that cannot elect a President has lost something more particular than a head of state. It has lost the office by which the other powers of the state are held in balance — the moderating power, the pouvoir neutre of the continental tradition, the arbiter that a parliamentary constitution installs precisely so that Government, Assembly, and Court cannot destroy one another through ordinary disagreement. On 28 April 2026 the Assembly of Kosovo was dissolved after it failed, for the third time in less than a decade, to elect that arbiter. The dissolution has been read as evidence of political failure — an intransigent opposition, a majoritarian government, a Court unable to compel compromise. This post offers a different framing. The dissolution was legal because it was automatic; it was automatic because the Constitutional Court, in a reading sustained since 2011, has made the vacancy of the presidency achievable by absence alone. The material stakes of Kosovar politics provided the incentive to use that mechanism; the mechanism itself is doctrinal. The office at the centre of both is the presidency, and the reason the presidency is at the centre is that the Court has, in more than one decision, diminished it.
The office at the centre
The presidency of Kosovo was designed as a moderating power — a figure of unity that stands above the three powers the Constitution recognises: Government, Assembly, and Court. The 2008 Constitution gives the office few direct powers of governance and many indirect powers of arbitration: representation of the state, nomination of the prime minister after elections, promulgation of laws, referrals to the Constitutional Court. The office is calibrated to help achieve consensus in matters of government formation and elections, precisely because its role is moderative rather than executive.
This calibration explains the presidential election procedure set out in Article 86 of the Kosovar Constitution. The first two rounds require 80 out of 120 votes; during the third round the approval threshold drops to 61, an absolute majority. This reduction is deliberate: the country must have a head of state, and if broad consensus proves unreachable, the Constitution provides for election by a narrower majority rather than for the office to stand empty.
What the Court has removed
However, the Constitutional Court has diminished the presidency of Kosovo in two distinct ways. In 2014, it narrowed what the office does.
Article 84 provides that the President appoints as prime minister the candidate proposed by the party or coalition holding the majority. Article 95, governing the same procedure, uses the past tense: the party that has won the election. One reading gives the President arbitrating authority — an assessment of which party can in fact form a government. The other reduces that assessment to simple arithmetic. In KO 103/14, the Court chose the second interpretation. The President was held to be bound by the certified electoral result, obliged to offer the mandate first to the largest single party regardless of coalition capacity. The arbitrating function of Article 84 was converted into a certifying one.
The second step taken by the Court relates to the reading of Article 86, which has governed presidential elections since 2011, and is doctrinally different. It does not reduce the powers of the office; it complicates the election process. Together, the two rulings leave the presidency less capable of serving as arbiter of the constitutional order — its powers narrowed when the office is filled, its filling obstructed when politics resists.
The reading the Court sustained
In 2011, in Sabri Hamiti and Others (KO 29/11), the Court invalidated the election of Behgjet Pacolli as President. In this judgment, the Court held that Article 86’s requirement of two-thirds of the votes of all deputies meant that all 120 deputies must vote, minus those with leave from the Speaker, with 80 affirmative votes required in the first two rounds. Fifteen years of institutional practice have softened this into an 80-deputy presence requirement: the Assembly is treated as unable to open a valid presidential session unless 80 deputies are physically present. An earlier post for this blog set out the doctrinal problems with this reading. What matters here is what the reading does. Under it, the third round of Article 86 — the safety valve that lowers the threshold when consensus fails — cannot be reached unless the earlier rounds were validly held, and they cannot be validly held without 80 deputies present. Failure of the presidential election, defined by Article 82 as the trigger for dissolution, becomes reachable by absence alone.
This is the constitutional mechanism as judicially re-interpreted. Whether it is used depends on politics.
The refusals
By March 2026, when the constitutional window opened, Vetëvendosje had a governing majority. The December 2025 snap elections had returned LVV to power, and a coalition including minority representatives — among them a Serb deputy — produced the 61 votes required to govern. What LVV did not have was the 80 votes required to elect a President in the first two rounds. The opposition parties PDK, LDK, and AAK commanded between them more than the 41 deputies whose absence would prevent an 80-deputy quorum under the Court’s reading.
The pattern of refusal to attend was not new. After the February 2025 elections, in which LVV had won a plurality but not a majority, Prime Minister Kurti proposed a coalition to the LDK; the LDK declined. The parliamentary session that followed produced a prolonged speakership crisis in which the Constitutional Court intervened repeatedly, structuring outcomes so tightly that ordinary political negotiation had little room to operate. No government was formed. Kosovo returned to the polls in December 2025. In the legislative period that followed, Kurti offered the LDK a joint government including four ministerial portfolios, one of them a deputy prime ministership and one of them a regalian ministry — foreign affairs. The LDK declined. Kurti offered the speakership of the Assembly to either PDK or LDK in exchange for their participation in the constitutive session. Both declined. Ahead of the presidential procedure of March 2026, Kurti made a further offer: the opposition parties could supply three candidates for the presidency of their own choosing, Vetëvendosje would provide the 30 signatures required for nomination and instruct its deputies to vote for one of the three. The opposition declined to supply candidates.
Taken together, these proposals described a pattern. Vetëvendosje was willing to trade specific institutional positions, up to senior cabinet portfolios and the choice of head of state, for coalition stability. The opposition declined every offer. The asymmetry is the empirical fact against which the incentive structure of the crisis needs to be read.
Both sides were acting on rational calculations, but their calculations were not symmetrical. LVV, having built its political identity on a critique of clientelism and state capture, could offer specific portfolios but not broad power-sharing that would compromise its programme — formalisation of the economy, minimum-wage increase, sovereignty enforcement in the north, anti-corruption prosecutions — and the offers made reflected that room. The opposition, facing a government whose programme would materially reduce the leverage of its supporting coalitions of interests, calculated that any participation in an LVV-led government would extend the government’s life and permit the consolidation it needed to prevent. The offers failed not because the terms were inadequate but because acceptance itself was the outcome the opposition had reason to prevent. Portfolios do not stop consolidation; only forcing a system reset does. The refusals were not tactical failures of negotiation. They were the substance of the opposition’s strategy.
The mechanism engaged
The presidency was the pressure point at which that strategy operated. President Vjosa Osmani sought a second term but could not secure the support required for nomination. As the deadline approached, she submitted constitutional amendments that would have replaced parliamentary election with direct popular election of the President; those amendments failed in the Assembly. The constitutional procedure then opened on the schedule Article 86 requires, and Vetëvendosje nominated Glauk Konjufca and Fatmire Mulhaxha-Kollçaku for the session of 5 March 2026 — procedural nominations placed on the ballot to satisfy the requirement that the process begin. The session failed for lack of quorum. Osmani issued a decree dissolving the Assembly on 6 March. The Constitutional Court invalidated the decree on 25 March in KO72/26 and KO74/26, giving the Assembly 34 more days.
Those 34 days confirmed the pattern the mechanism made available. The compromise offer already described was rejected. On 27 April, the day before the deadline expired, Vetëvendosje nominated Feride Rushiti and Hatixhe Hoxha and called an extraordinary session for the same evening. 64 deputies were present. A vote was held. Rushiti received 63 votes. Under the Constitution’s design, this was a validly held round that failed and should have advanced the procedure. The Court’s presence requirement prevented that advancement. The session was interrupted and resumed on 28 April, but no further rounds took place. The Assembly dissolved at midnight when the deadline expired. Snap elections followed on 7 June and returned an Assembly with the same fundamental configuration.
What the equilibrium requires
Kosovo’s dissolution cycle is sustained by an institutional architecture that made a specific tactical option available and by a political situation in which that option is rational for the actors who possess it. The Court’s reading of Article 86 supplied the mechanism. The gap between LVV’s governing majority and the 80 votes the Court’s reading requires supplied the occasion. The material stakes of LVV’s programme supplied the motive. An office designed to hold the three powers in balance has become the instrument through which one of them is periodically destroyed. The arbitrating powers of the presidency have been narrowed by the Court, and the office itself has become progressively harder to fill. The pouvoir neutre is not merely absent in the sense of being unelected. It is absent in the deeper sense that even when elected, it has been given less to do.
The way forward
Constitutional amendment is not currently available. Amending Article 86 to remove the interpretive supplement would require a two-thirds majority in the Assembly — the same threshold whose current reading has produced the crisis.
The path that remains is interpretive. The Constitutional Court retains the authority to state, prospectively, that Article 69’s quorum of 61 deputies governs every session of the Assembly, including the session that elects the President; that Article 86’s thresholds are winning numbers rather than presence requirements; and that a round of voting held under the Article 69 quorum in which no candidate reaches the threshold is a round that has occurred and failed, triggering the next stage — the next round, the third round, or, after a third round in which 61 affirmative votes have not been gathered, the dissolution Article 82 provides. This correction does not overrule the 2011 case; it distinguishes the holding of that case from the reasoning offered after the case was already decided. It restores the architecture the Constitution wrote for the next presidential election and the legislatures that follow.
The office designed to arbitrate a democratic republic cannot survive if its vacancy has become the default consequence of ordinary partisan disagreement. And the stake is not the presidency alone. What is at risk is the capacity to translate an election result into a working government. When procedural absence effectively becomes a veto of the outcome of an election, the constitutional order stops recognising the electorate’s decision and instead rewards the minority that refused to participate in it. Restoring the presidency to its constitutional role is therefore not a partisan project. It is the precondition for the country having a head of state at all — and for elections to mean what they are supposed to mean. A constitution interpreted to reward its own paralysis ceases to be a framework for governance and becomes a blueprint for institutional collapse. Until the Court closes the mechanism its reading has kept open, the pattern will repeat, and each repetition will confirm what the last one demonstrated — that in Kosovo, electoral majorities are provisional and dissolution is on offer to whichever minority chooses to invoke it.
Suggested citation: Gazmend Demolli, The Narrowed Office: How Kosovo’s Constitutional Court Has Diminished the Presidency and Made Dissolution a Political Tool, Int’l J. Const. L. Blog, Aug. 8, 2026, at: http://www.iconnectblog.com/the-narrowed-office-how-kosovos-constitutional-court-has-diminished-the-presidency-and-made-dissolution-a-political-tool/