Update
How CRTA observes elections: inside the methodology and stru...
Standards, methodology and structure of the mission CRTA deploys for the 2026 elections
The OSCE Office for Democratic Institutions and Human Rights (ODIHR) opened its election observation mission in Belgrade on 16 September 2026, for the early parliamentary elections called for 25 October. The mission is headed by Eoghan Murphy and consists of a core team of 14 international experts based in Belgrade. Thirty long-term observers, drawn from 24 OSCE participating States, deploy across the country on 25 September, one month before election day, and ODIHR has requested that participating States second 300 short-term observers to arrive several days before the vote.
The mission will follow the campaign, including online and social media, campaign financing, the work of the election administration, the legal framework and its application, the arrangements in place to detect and counter disinformation, media coverage of the campaign, and the resolution of electoral disputes. It will also assess the implementation of previous ODIHR recommendations. "Elections are a process, not a single day, and our observation begins well before voters go to the polls," Murphy said at the opening, adding that the assessment "will be independent and impartial, and grounded in ODIHR's established methodology and the commitments and standards that apply across the OSCE region".
The mission delivers what ODIHR recommended in July. Its Needs Assessment Mission report of 30 July called for a full Election Observation Mission with a core team of analysts, long-term observers across the country and 300 short-term observers for election day - the largest mission ODIHR had proposed for Serbia in over a decade. The comparison with earlier missions bears that out: for the December 2023 early parliamentary elections ODIHR fielded a core team of 11 experts, 30 long-term observers and 250 requested short-term observers; for the April 2022 elections, 15 core experts, 27 long-term observers and 250 short-term observers. The increase is in the core team and, above all, in election-day coverage, where 300 short-term observers represent a fifth more than three years ago - the area interlocutors flagged most sharply after the irregularities and pseudo-observers recorded at the local elections in March 2026.
The mission's own calendar will shape much of the public debate in the final month of the campaign. ODIHR will publish an interim report approximately two weeks before election day, present a statement of preliminary findings and conclusions at a joint press conference the day after the vote, and issue a final report with recommendations a few months later. That sequence runs alongside the domestic electoral calendar: applications for accreditation by international observers close on 14 October and by domestic observers on 17 October, so the composition of the observation effort as a whole will be settled before the interim report appears.
For CRTA, the substance of this mission lies less in its size than in the standard it applies. Serbia's electoral framework was amended in the run-up to these elections, and the authorities present those amendments as the answer to years of criticism. ODIHR's explicit mandate to assess the implementation of previous recommendations puts that claim where it belongs - against the record. CRTA's own assessment found that the recommendations addressed so far are largely technical, while those touching the level playing field - the misuse of administrative resources, pressure on voters, campaign finance oversight, media balance, dissuasive sanctions - remain open. What the mission measures is not whether recommendations were adopted, but whether they are applied.
International observation also does not stand on its own. ODIHR's long-term observers will cover the country from 25 September with 30 people; the density of coverage on election day, and the continuity of monitoring in the years between elections, come from domestic observers. CRTA will again deploy its own election observation mission, and its findings, together with those of other domestic organisations, will form part of the evidence available to the international mission. Furthermore, a robust observation presence raises the cost of manipulation without removing its causes. Whether these elections are credible will be decided by how the rules are applied over the next five weeks, not by how many observers were in the country to watch it happen.
The mission will follow the campaign, including online and social media, campaign financing, the work of the election administration, the legal framework and its application, the arrangements in place to detect and counter disinformation, media coverage of the campaign, and the resolution of electoral disputes. It will also assess the implementation of previous ODIHR recommendations. "Elections are a process, not a single day, and our observation begins well before voters go to the polls," Murphy said at the opening, adding that the assessment "will be independent and impartial, and grounded in ODIHR's established methodology and the commitments and standards that apply across the OSCE region".
The mission delivers what ODIHR recommended in July. Its Needs Assessment Mission report of 30 July called for a full Election Observation Mission with a core team of analysts, long-term observers across the country and 300 short-term observers for election day - the largest mission ODIHR had proposed for Serbia in over a decade. The comparison with earlier missions bears that out: for the December 2023 early parliamentary elections ODIHR fielded a core team of 11 experts, 30 long-term observers and 250 requested short-term observers; for the April 2022 elections, 15 core experts, 27 long-term observers and 250 short-term observers. The increase is in the core team and, above all, in election-day coverage, where 300 short-term observers represent a fifth more than three years ago - the area interlocutors flagged most sharply after the irregularities and pseudo-observers recorded at the local elections in March 2026.
The mission's own calendar will shape much of the public debate in the final month of the campaign. ODIHR will publish an interim report approximately two weeks before election day, present a statement of preliminary findings and conclusions at a joint press conference the day after the vote, and issue a final report with recommendations a few months later. That sequence runs alongside the domestic electoral calendar: applications for accreditation by international observers close on 14 October and by domestic observers on 17 October, so the composition of the observation effort as a whole will be settled before the interim report appears.
For CRTA, the substance of this mission lies less in its size than in the standard it applies. Serbia's electoral framework was amended in the run-up to these elections, and the authorities present those amendments as the answer to years of criticism. ODIHR's explicit mandate to assess the implementation of previous recommendations puts that claim where it belongs - against the record. CRTA's own assessment found that the recommendations addressed so far are largely technical, while those touching the level playing field - the misuse of administrative resources, pressure on voters, campaign finance oversight, media balance, dissuasive sanctions - remain open. What the mission measures is not whether recommendations were adopted, but whether they are applied.
What the mission measures is not whether recommendations were adopted, but whether they are applied.
International observation also does not stand on its own. ODIHR's long-term observers will cover the country from 25 September with 30 people; the density of coverage on election day, and the continuity of monitoring in the years between elections, come from domestic observers. CRTA will again deploy its own election observation mission, and its findings, together with those of other domestic organisations, will form part of the evidence available to the international mission. Furthermore, a robust observation presence raises the cost of manipulation without removing its causes. Whether these elections are credible will be decided by how the rules are applied over the next five weeks, not by how many observers were in the country to watch it happen.
