The Count That Doesn’t Add Up

ETHIOPIAN TRIBUNE Election Analysis — June 2026

The Count That Doesn’t Add Up

Ten days after Ethiopians went to the polls, the National Election Board is still totting up the figures, the system that produces such lopsided results remains exactly as it was, and a flank of the opposition has pre-emptively disowned a parliament it may yet have the numbers to sit in.

By E Frashie Ethiopian Tribune Columnist

Ethiopia’s seventh general election was, by the standards its own organisers set, a triumph of scale. Over 50 million registered voters; more than 10,900 candidates from 47 parties; a national holiday declared to encourage turnout. By the standards anyone else might apply, it has been a triumph of something else entirely: the art of taking a very long time to arrive at a very predictable answer.

Voting closed on 1 June. By NEBE’s own admission that evening, counting was still under way in Sidama, Gambella, Amhara and Somali, and final turnout figures had not been consolidated. On 6 June, the Board reported that 825 of 1,138 constituencies had declared results, citing the distance between polling stations and constituency centres and the sheer number of candidates as reasons for the lag. On 9 June  a full week after polling day internally displaced people and military personnel cast their votes in a special round that, by NEBE’s own insiders, threatened to delay preliminary results still further. By 10 June, 1,008 constituencies had reached the national verification centre, with 446 of 501 House of Peoples’ Representatives seats and 562 of 638 regional council seats logged for “final scrutiny.” A day later, both NEBE and the African Union observer mission had pencilled in 11 June for the announcement of final results.

None of this should be read as evidence of unusual incompetence. It is, if anything, on brand. In 2005, results were delayed for weeks while complaints of vote-rigging were investigated investigations the opposition’s own representatives called “rigged from the start,” alleging their observers had been “harassed, threatened, barred and killed.” In 2010, the electoral board rejected calls for a re-run despite the ruling coalition and its allies taking 534 of 537 seats, a margin opposition leaders said “could not be accomplished without cheating.” The pattern across two decades is not that Ethiopia’s election commission is slow because it is overwhelmed. It is slow because slowness is the institutional environment in which adjustments, where they occur, are made  and because nobody with the power to speed it up has ever had an interest in doing so.

THE MODERNISATION THAT WASN’T

What makes the 2026 cycle different is the sheer quantity of money spent trying to fix exactly this problem. The “Strengthening Ethiopia’s Elections for Democratic Sustainability” programme (SEEDS2) a multi-donor basket funded by Austria, Denmark, Ireland, Spain, the United Kingdom and the Government of Ethiopia, administered through the UNDP  was built explicitly around “electoral digitalisation.” UNDP signed a $40 million project document with NEBE to build “institutional and logistical capacity.” Japan contributed a further $3 million in equipment and systems “for the upcoming 7th General Election.” Ireland signed a separate €700,000 agreement aimed specifically at helping NEBE “adopt new technologies to lead a transparent and efficient electoral process.” The EU and Germany, via the KfW development bank, funded a parallel programme EURECS+ delivered jointly with UNDP and the International Foundation for Electoral Systems, providing “further logistical and procurement support” to NEBE’s operations.

The visible product of all this is Mirchaye (ምርጫዬ) the digital platform NEBE unveiled in January 2026 described at launch as “the first time the country is fully integrating digital technology into the election process,” covering voter and candidate registration, election management, and, in the words of the Federal Supreme Court’s president, present at the launch, intended to “enhance transparency and public trust.”

What Mirchaye (ምርጫዬ) did not touch, it turns out, is the part of the process that actually generates disputes. Registration, candidate vetting, the ballot-order lottery  all digitalised, all relatively uncontroversial. Counting, constituency-level tabulation, and the transfer of results to Addis Ababa for “verification” all still conducted in essentially the form they took twenty years ago: paper postings at polling stations, physical transmission to constituency offices, and a national tally centre that receives results in batches over a period of days, occasionally weeks. Opposition parties identified this gap before the vote was even held. In a joint statement issued in January, an eight-party coalition  including the Ethiopian People’s Revolutionary Party (EPRP), Hibir Ethiopia and the Wolayta National Movement rejected what they called an “unaudited digital election system,” accusing the Board of “uncoordinated increases in council seats and a lack of transparency.”

There are two ways to read this gap, and they are not mutually exclusive. The first is straightforwardly bureaucratic: a decade of donor funding produced visible, demonstrable, photographable modernisation  apps, registration drives, ballot lotteries  because those are the outputs donors can point to in their own annual reports, while the unglamorous, politically fraught work of building an auditable, end-to-end digital chain of custody for actual vote counts was either never prioritised or actively resisted by an institution with no incentive to make its own discretion harder to exercise. The second reading is less generous: that a tabulation process kept manual, opaque and slow is not a failure of modernisation but its precondition. An auditable digital count would generate a data trail far harder to adjust at the margins than a process whose authoritative record exists only in the moment a regional official reads a number aloud to a colleague in Addis Ababa. Either way, the tens of millions of dollars spent since 2019 modernising Ethiopia’s elections have left the single most consequential five days of the electoral calendar the count  almost exactly as they found it.

THE ARITHMETIC THAT NEVER CHANGES

None of this would matter nearly as much under a different electoral system. Ethiopia’s first-past-the-post arrangement, inherited from the 1995 constitution and never seriously revisited despite two decades of promises, converts even modest vote margins into landslide seat counts. In 2021, the Prosperity Party took 96.8% of federal parliamentary seats from roughly 90% of the vote itself a serious overcorrection, but nothing compared to the distortions at constituency level. In Addis Ababa that same year, opposition parties collectively won 32% of the vote and not a single seat. In 2015, the Semayawi Party took 16% of the vote in the capital and, again, nothing. In 2005, opposition parties won around 38% of the national vote and ended up with a small fraction of that share in seats.

The reform conversation is not new, and it has never gone anywhere. After the unrest of 2016, the government promised to revisit the electoral system; no measure followed. Years earlier, during the EPRDF-era political parties’ negotiation forum, the ruling party itself proposed a mixed system, 90% first-past-the-post, 10% proportional which would have required expanding the House of Peoples’ Representatives from 550 to 657 seats and amending the constitution. Even that modest concession was shelved. Academic studies of the system, going back to at least 2017, have concluded that first-past-the-post is “ill-suited to Ethiopia’s current needs and realities” and that a shift toward proportional representation the system used by roughly 130 countries, against fewer than 55 still using FPTP would produce more inclusive, more stable outcomes even under identical levels of political repression.

The reason none of this has moved is not mysterious. A ruling party that can convert a plurality into a supermajority under FPTP has no rational incentive to adopt a system that would convert the same plurality into, at best, a comfortable majority and, at worst, a coalition negotiation. Proportional representation would not, on its own, make Ethiopian elections free or fair, repression, detention and restricted media access would still shape who could compete. But it would make the result of that repression visible in a way FPTP currently launders into invisibility: a Prosperity Party that wins 60% of the vote under PR gets roughly 60% of the seats, prompting awkward questions about the other 40%. Under FPTP, that same 60% becomes 95% of the seats, and the other 40% becomes a rounding error. Donor money has, for fifteen years, flowed toward the parts of Ethiopia’s electoral architecture that are easiest to fund and least likely to threaten this arrangement. The one structural change that would actually rebalance outcomes was never going to be financed by the institution whose entire utility to the ruling coalition depends on its absence.

“WE CAME TO DOCUMENT, NOT TO WIN”

It is against this backdrop that the position taken by the Coalition for Ethiopian Unity becomes legible not as an eccentric or self-defeating gesture, but as possibly the most coherent response on offer. In an interview given after the vote, Mister Silassie Tamerat, Secretary General of both the Coalition and the EPRP, laid out a position that amounts to participation without consent: contesting the election while refusing, in advance, to recognise its outcome.

The Coalition and the EPRP had set out roughly seven preconditions before the vote chiefly concerning media independence, the release of political prisoners, and an end to the fighting in Tigray, Amhara and Oromia. “Absolutely none” were met, Tamerat said; if anything, conditions deteriorated, with media outlets shut down, journalists abducted, and election debates censored or cut from broadcast. The decision to participate anyway was not unanimous. Within the EPRP’s Central Committee, a majority judged that continued participation “exposing the fraudulent electoral process from within” while “protecting the security and survival of our party organisation” outweighed the principle of boycott. MistreSilasie by her own account, was outvoted, and deferred to the decision as a matter of internal democratic discipline.

What followed the vote was the more consequential decision: a declaration that any Coalition or EPRP candidates who won seats would not take them up. The reasoning rests on a distinction between participating in an election and recognising a parliament.

“Entering a parliament that is born from a completely fraudulent and illegitimate process would imply that we recognise the parliament as legally and legitimately established. We do not believe that.”

— MistreSilasie Tamerat, Secretary General, Coalition for Ethiopian Unity / EPRP

She pointed to the precedent of past opposition MPs “stripped of their immunity and thrown into prison simply because they held a dissenting political viewpoint” as evidence that the inside-the-system strategy had already been tried, and had “brought no real change.” Sitting in the chamber and “declaring our people are suffering,” in her words, “does not bring change.”

Asked whether refusing seats amounted to disrespecting the voters who cast ballots for Coalition candidates, Mistre-Silassie’s answer was unambiguous about priorities, if not entirely reassuring on the question of mandate: the Coalition, she said, had announced this position before polling day “even if we won a majority, we had no desire to enter parliament under these conditions” and had entered the race “merely to gather hard evidence,” not to contest for seats. Critics who suggest the rejection is sour grapes over an anticipated loss get short shrift: “We did not enter this process expecting to win… This has nothing to do with winning or losing.”

A COALITION OF ONE MIND, MOSTLY

The recent interview of Mistre-Silassie  with Ethiopian insider also surfaced and sought to manage reports of a split within the Coalition. The president of the Amhara Regional United Movement, one constituent member, had told the press the election had been conducted “smoothly and in a good manner” an assessment, Mistre-Silassie noted pointedly, that was “heavily amplified” by state-affiliated media. She was unambiguous in disowning the remark: the individual “was not authorised or delegated by the Coalition,” the Coalition’s “core principles” remained intact, and the organisation was “internally purifying” its ranks rather than fracturing. Any elected candidate who attempted to take a seat under the Coalition’s banner without its blessing, she said, would face “immediate” disciplinary and legal consequences.

It is worth taking at face value Mistre-Silassie’s own account of what worries her most  because it is not, by her telling, the regime.

“Dealing with the tyranny of the ruling regime is something we expected. What deeply concerns and threatens to break my hope is the severe fragmentation within the opposition itself — fractured, petty, and constantly turning on one another.”

 Mistre-Silassie Tamerat

A planned peaceful protest for 30 April for which, under the constitution, only notification rather than permission was required was shelved for “logistical pressures and constraints” after a decision to act collectively rather than unilaterally; it never materialised, nor did any other rally. The EPRP, Mistre-Silassie said, held back its own planned action “to respect the principle of collective coalition solidarity.” The through-line across both the Amhara Regional United Movement episode and the failed protests is the same: a coalition whose member parties cannot reliably act in concert is, for practical purposes, several smaller oppositions wearing one name a condition that costs the ruling coalition nothing and the opposition everything.

THE VERDICT ON THE VERDICT

On the international response, Mistre-Silassie’s framing inverts the usual Mistre-Silassie courtesies. Any observer mission that grants the election legitimacy, she argued, is “actively participating in prolonging the misery and suffering of the Ethiopian people.” She drew a distinction between observers who acknowledged the vote merely “took place” reserving final judgment, as the EU did, for a longer review  and any body offering outright congratulations, noting with something close to satisfaction that many embassies briefed by the Coalition on election day had so far stayed silent. “We view that silence as a major diplomatic achievement for our cause.”

She was careful, notably, not to indict NEBE’s staff wholesale acknowledging “certain individual officials” who “genuinely tried their best to ensure a clean process” while insisting the institution as a whole had failed systemically, operating, like every other institution in the country, under “the suffocating influence and control of the ruling regime.” The war is the backdrop against which she places the entire exercise: Tigray “plunged into conflict for a second time,” most of Amhara unable to vote due to active combat, Oromia’s security situation outside major cities “non-existent.” Against that, talk of “free and fair” becomes, in her telling, close to obscene: “citizens are actively dying in a war zone and screaming for the killings to stop rather than asking for voting cards.”

On armed resistance, Tamerat drew a careful line denying coordination with armed groups while declining to condemn them, and issuing what reads as a warning dressed as an observation: “Every single time you violently close the doors to peaceful political struggle, you automatically open the doors to armed resistance… No one picks up a gun and flees to the jungle to die or kill out of choice; they do it out of absolute desperation when all peaceful options are stripped away.”

THE SYSTEM’S MOST DURABLE OUTPUT

Put the three threads together and a single picture emerges, and it is not primarily a picture of fraud in the narrow sense ballot-stuffing, doctored tallies, the things that require forensic proof. It is a picture of a system whose component parts are individually defensible and collectively self-reinforcing. The count is slow because the institution that runs it has never had to make it fast: under first-past-the-post, even a contested, drawn-out tabulation poses no existential risk to the ruling coalition’s seat total, so there is no urgency to fix what donors have spent fifteen years and tens of millions of dollars failing to fix because the fix that would matter, electoral reform, was never on the table to begin with. And the opposition’s response to all of this is not unified resistance but fragmentation: a Coalition that agreed on seven preconditions, watched none of them met, then split three ways over what to do next participate to expose, boycott to preserve legal standing, or some uneasy combination of both, followed by a post-hoc refusal to take the seats that participation might have won.

That fragmentation is not incidental to the system described above. It may be its most reliable product. A ruling party facing a unified opposition demanding proportional representation, an end to detentions, and international observers with teeth would face a genuine cost-benefit calculation. A ruling party facing a half-dozen opposition factions arguing among themselves about whether occupying five parliamentary seats constitutes principled resistance or collaboration faces no such calculation at all. Tamerat’s own diagnosis that internal opposition fragmentation, not regime pressure, is what “keeps her up at night”  may be the most analytically honest sentence to emerge from Ethiopia’s seventh general election. Everything else, the slow count included, was entirely predictable.

THE REGION THE ELECTION FORGOT

For TPLF supporters, the chronology of NEBE’s dealings with their party reads less like routine regulation than personal score-settling. Birtukan Mideksa rose to prominence as a leader of the Coalition for Unity and Democracy in the disputed 2005 election  conducted under an EPRDF government in which the TPLF was the dominant faction and was subsequently imprisoned, controversially re-arrested after a conditional pardon, and spent years in American exile before Abiy brought her home in 2018 to chair the very board that had once certified the election that led to her imprisonment. When NEBE, under her chairmanship, cancelled the TPLF’s registration in January 2021  weeks after the federal offensive into Tigray began TPLF sympathisers read the timing as more than coincidence: a woman the old TPLF-dominated order had jailed was now the one stripping that same party of its legal existence. The institutional posture has since outlasted her TPLF’s final de-registration came in May 2025 under her successor, Melatework Hailu but the original grievance, whatever its merits, has never gone away in Mekelle.

Its consequences are visible in real time. Tigray did not participate in the 1 June election at all: Debretsion Gebremichael announced in May that no vote would be held in the region, arguing large parts of it remain outside TPLF administration and that Tigray’s territory is “incomplete” pending unresolved disputes  itself a legacy of NEBE’s 2021 decision and the deregistration that followed. While NEBE in Addis Ababa spent the second week of June finalising a result that hands the Prosperity Party its expected supermajority, a very different kind of delegation was landing in Mekelle. On 11 June, the African Union’s Olusegun Obasanjo one of the architects of the 2022 Pretoria Agreement  flew in with British and Australian diplomats for emergency talks with Debretsion, who has installed himself as Tigray’s regional president following the TPLF’s unilateral revival of the pre-war regional council, a move the federal government has refused to recognise.

The same day, in Al Jazeera, Getachew Reda  the former TPLF spokesman who broke with the party, led the interim administration, and now advises Abiy on regional affairs co-wrote with the federal government’s own Pretoria negotiator, Redwan Hussein, a piece warning that the “rump TPLF,” in what they called an Eritrean-backed “Tsimdo alliance,” had “openly abrogated the Pretoria Agreement and is now gearing up for active and open hostility against the federal government.” Whatever one makes of that framing  and it is very much the view from the side that now controls Addis Ababa’s narrative on Tigray, not TPLF’s  the fact that the AU’s most senior mediator felt the need to be in Mekelle that same week speaks for itself.

The result is a grim symmetry. Ethiopia just completed an election in which the region with the most unresolved grievance against the federal state didn’t vote at all, conducted by an institution whose history with that region’s dominant party is, by the account of that party’s supporters, inseparable from the personal history of the woman who once ran it. The seventh general election may produce a parliament. Whether it produces a country at peace with itself by the time that parliament sits is, as of this week, genuinely an open question.