How many of the negotiating chapters required to join the EU will Montenegro have closed at the end of 2026?

TL;DR

My forecast centers on 29 provisionally closed chapters at the deadline; the mean is 28.0, the median is 29, and 30 is the single most likely exact result. I assign 40% to 30 or more, 16% to 32 or more, and 8% to all 33. Montenegro closes most technical files but carries one or more rule-of-law, environmental, budgetary, or Croatia-sensitive chapters into 2027.

Context

The client's March baseline is stale. Montenegro formally closed Chapters 8 and 29 on 14 July 2026, bringing the official count to 18 of 33; the 15 still open are Chapters 1, 9, 12, 14–19, 22–24, 27, 31 and 33 (Council status page). Reopening an already closed chapter is possible but has no current evidentiary support, so 18 is the practical floor.

Formal closure needs a favorable Commission assessment, an EU common position accepted unanimously by the member states, and confirmation at an accession conference (European Commission process guide). The remaining calendar contains four plausible formal General Affairs Council windows—22 September, 13 October, 19 November and 15 December—so the process has enough meeting dates, but not enough room for repeated delays.

Evidence

The historical backbone is Montenegro's complete closure record. The process was nearly dormant until the rule-of-law interim benchmarks were cleared in 2024, then accelerated sharply.

Croatia is the closest completed reference class. It moved from 28 of 35 chapters closed on 22 December 2010 to all 35 on 30 June 2011. That was seven closures in just over six months, but one was procedural Chapter 35, leaving six substantive policy chapters. Montenegro needs 15 substantive chapters in roughly four and a half months. The analogue makes a large final batch credible, but makes a clean sweep a tail outcome.

The current case is stronger than a simple cadence extrapolation suggests. Ireland has made completion a presidency priority and publicly committed to facilitating the year-end target (Irish government, 27 January 2026). The Council created a working party to draft Montenegro's accession treaty on 22 April 2026, and the Commission presented the Chapter 33 financial package on 30 June 2026. These are real endgame steps, not generic diplomatic praise.

Several individual chapters have useful leading indicators. Montenegro says Chapter 14 is technically ready and expects a September conference (24 July 2026). For Chapter 18, the government adopted an ESA 2010 roadmap whose content the Commission had approved as addressing the remaining benchmark (9 July 2026). For Chapter 27, Montenegro submitted eight laws, several strategies and more than 95 secondary acts by 15 March 2026, though administrative capacity and transition terms remain hard.

The downside is concentrated rather than broad. The Commission's June rule-of-law non-paper said Chapters 23 and 24 were progressing but still required work on judicial appointments, corruption and organized-crime results, visa policy, border management, media freedom and fundamental rights (11 June 2026 summary). Croatia's foreign ministry then said on 22 July 2026 that bilateral disputes must be resolved to complete the negotiations. The list includes wartime compensation, missing persons, prosecutions, property claims, the Morinj memorial, the Kotor pool name, the maritime boundary and the training ship Jadran. The more positive signal is that the two governments agreed in August to hold focused talks on the maritime boundary and Jadran (8 August 2026), which looks more like leverage-seeking than a permanent veto.

The domestic workload is still large. Montenegro's accession programme identified 122 closure-critical obligations: 22 strategic documents, 48 laws and 52 secondary acts (government programme briefing). The programme shows serious mobilization, but it also explains why technical readiness can slip even when political support is strong.

I modelled the 15 open chapters in four groups: seven relatively advanced or unblocked chapters, four harder technical chapters, two Croatia-exposed chapters, and the two rule-of-law chapters. I mixed five correlated scenarios rather than treating every chapter as an independent coin flip.

The full-package scenario is not the only route to 33, so the resulting probability of exactly 33 is 8%. The final distribution has a mean of 28.0, median 29, mode 30 and a central 80% interval of 23–32 chapters.

What's non-obvious

The Croatia comparison cuts both ways. It proves that rule-of-law, competition and budget chapters can close together in one final conference. But Croatia closed only six substantive chapters during its last six months, while Montenegro must close 15 before New Year's Eve. Headlines that cite Croatia as proof that all 33 are on track miss that arithmetic (Croatian endgame record).

Treaty drafting and the financial package raise the odds of a high total, but they do not create a hard December deadline. Commissioner Marta Kos explicitly said that spilling some closures beyond 2026 would not automatically prevent membership in 2028 (14 July 2026). That lowers the incentive to force through Chapters 23, 24, 27 or 31 before the evidence and diplomacy are ready. It is the main reason the forecast stops near 29 rather than treating 33 as the base case.

Uncertainties

• The Council has not publicly released a chapter-by-chapter pipeline showing which draft common positions have cleared Commission review, COELA or COREPER. That information would sharply narrow the distribution.
• Montenegro has announced a September conference for Chapter 14, but it was not yet listed as a confirmed accession conference on the Council calendar at the cutoff.
• Croatia's minimum acceptable settlement is private. A roadmap or arbitration commitment could unlock Chapters 14 and 31 quickly; a demand for final settlement of every issue could push both into 2027.
• The decisive technical gaps are the Commission's final benchmark assessments for Chapters 23 and 24 and the transition-period agreement for Chapter 27. Public statements show movement, but not whether all member states will accept closure by December.