There is a simple paradox at the heart of every credible democratic election: there should be procedural certainty but uncertainty about the outcome. The rules must be clear before the contest begins, the referee must be trusted by all sides, citizens must be free to campaign, organise and vote and the counting process must be transparent and traceable. What should remain uncertain until the last legitimate vote is counted is who will win.
It is against that standard that Zambia's 2026 election should be judged.
I have observed every Zambian presidential election since 2006, when Hakainde Hichilema first contested the presidency. I have therefore seen the country move through successive electoral cycles, including elections marked by intense competition and disputed results.
I was also present at the national results centre at Mulungushi International Conference Centre after completing my field observation. My concern about the 2026 election is therefore not that Zambia experienced a difficult election. It is that several institutional safeguards appear to have weakened simultaneously, creating legitimate questions about whether the election provided procedural certainty and genuine uncertainty of outcome.
The immediate cure is not another political argument. It is evidence. Zambia should commission an independent forensic audit of the results chain, beginning with the Gen-20 forms at polling stations and following them through constituency and district aggregation to the national results centre.
Every polling station produces documentary evidence. As Kenya demonstrated through the use of independent firms such as KPMG to scrutinise electoral data, Zambia could similarly engage a reputable international firm, KPMG, PwC, Deloitte or EY, to independently verify the integrity and chain of custody of the results.
The question is simple: Can the declared national result be reconstructed from the votes recorded at polling stations? Political parties and local monitors should have copies, the ECZ should have copies and the figures should reconcile. If they do, an audit strengthens the declared result. If they do not, the discrepancies must be identified and investigated. The question is not whether Brian Mundubile of the National Reconciliation Party for Unity and Prosperity (NRPUP/Tonse Alliance) or Hakainde Hichilema of the United Party for National Development Alliance (UPND) won. The question is whether the declared result can be reconstructed from the votes cast at Zambia's polling stations. The question has become necessary because the problems with this election did not begin on polling day.
**A deteriorating political environment **
The pre-electoral environment was increasingly hostile to democratic competition. Human Rights Watch and Amnesty International documented concerns over restrictions on civic space, on and offline, freedom of expression and assembly, the treatment of journalists and opposition actors and the use of legislation, such as the Cyber Crimes Act and Cyber Security Act, in ways that created a chilling effect on dissent. These are not merely human-rights concerns. They are electoral concerns.
An election is not simply the act of placing a ballot in a box. Opposition parties must be able to organise and campaign; journalists must be able to investigate; civil society must be able to observe; and citizens must be able to criticise government without fear. If the freedoms are progressively narrowed in the months before voting, the existence of a ballot does not restore a level playing field.
Local civil-society organisations raised similar concerns. There were allegations of inducements to opposition candidates to withdraw, vandalism of campaign materials, restrictions on opposition meetings and violence affecting opposition activities. Opposition parties also faced difficulties campaigning in Southern Province.
Whatever the political history of the province, an opposition candidate must be able to campaign there as freely as the incumbent's party. The European Union (EU) Election Observation Mission subsequently described the election as competitive but concluded that restrictions on fundamental freedoms contributed to an uneven playing field. The finding should not be casually dismissed.
Weakening the opposition
The weakening and fragmentation of the opposition was another defining feature of the election. After Edgar Lungu's death, the Patriotic Front (PF) entered the contest divided, with the struggle over who controlled the party becoming entangled with the broader political contest. Robert Chabinga emerged as leader of one PF faction and subsequently aligned himself with Hichilema, amid allegations that the fragmentation was facilitated by the state to prevent the president from facing a credible and consolidated challenger.
There is a striking regional parallel with Sengezo Tshabangu's role in Zimbabwe's Citizens Coalition for Change (CCC), where an individual outside the established leadership claimed control of the party and triggered the recall of opposition legislators. The cases are not identical but the political effect is comparable: weakening the opposition not by defeating it at the ballot box but by disrupting the organisational structures through which it contests power. By the time Brian Mundubile emerged as the principal opposition candidate, only three months or so before the election, the opposition had lost valuable political ground.
The opposition has alleged that the fragmentation was not simply an internal PF problem but was facilitated by the state because the incumbent wanted to avoid facing a credible and consolidated opposition. That is a serious allegation and should not be presented as proved without evidence. But the political outcome is beyond dispute: the traditional opposition entered the election weakened, divided and forced into new political arrangements.
Brian Mundubile's emergence as the principal opposition challenger came only months before polling. The EU mission noted that he became the principal opposition candidate only three months or so before the election after the formation of the Tonse Alliance. The point is not that opposition parties have an entitlement to win. They do not. The point is that democratic competition requires more than formal permission to contest; it requires a reasonable opportunity to become politically viable.
The Registrar of Societies became particularly important in this regard. Opposition parties complained of hostility and administrative obstacles, including difficulties around party registration and changes in leadership. Some political parties were reportedly registered or treated as churches. The allegations require documentary verification but the underlying institutional concern is serious: an administrative office capable of determining whether a political organisation legally exists can become a consequential actor in an election without ever appearing on a ballot.
**Constitutional reform and the boundaries of competition **
Constitutional Amendment No. 7 presents a more complicated picture. There are genuine democratic gains in the reform, particularly the introduction of mixed-member proportional representation and additional parliamentary seats for women, youth and persons with disabilities.
The changes should be acknowledged. But the amendment also increased constituency seats from 156 to 226, changing the architecture of Parliament. The creation of new constituencies generated accusations of gerrymandering. The allegation should not simply be repeated as fact; it requires analysis of population distribution, voter numbers, geography and historical voting patterns. But constituency boundaries are never politically neutral. They determine how votes become seats.
The African Union-Comesa and other observer missions raised concerns about the delimitation process and the limited time available for voter education around the new boundaries. The issue therefore deserves independent scrutiny, particularly because constitutional and electoral reforms of this magnitude should command confidence across the political spectrum.
**The referee problem **
The most consequential institutional issue is the Electoral Commission of Zambia (ECZ). An electoral commission does not become credible simply because its appointments comply with the law. It becomes credible when contestants believe it is independent and when its conduct demonstrates that independence.
The ECZ leadership entered the election carrying substantial controversy. Chairperson Mwangala Zaloumis had served as Hichilema's personal lawyer. Commissioner McDonald Chipenzi has been publicly associated with the ruling United Party for National Development (UPND), while Zevyanji Sinkala's previous political associations have also generated controversy. The circumstances do not prove that any commissioner acted improperly. But they create an obvious perception-of-independence problem, particularly when the commission later makes decisions that affect the incumbent's electoral prospects.
That problem was compounded by the appointment of Maurice Hichilema, the president's brother, as secretary of the Local Government Service Commission on the eve of the election. The strongest objection is not simply that he is the president's brother. It is the institutional role of the office. Local government structures are embedded in the administrative architecture through which electoral information and results move. Town clerks and other local-government officials play important roles at district level, including in the broader results-management process. The appointment therefore created an avoidable conflict of perception at precisely the point when institutional neutrality mattered most. The issue is not whether Maurice Hichilema is competent. It is whether an administration concerned about public confidence should knowingly create such a politically sensitive institutional arrangement during an election.
**The securitisation of the election **
The securitisation of the electoral process was another major red flag.The replacement of Army Commander Lieutenant General Sitali Alibuzwi by Geoffrey Zyeele was politically controversial, particularly given Zyeele's previous relationship with Hichilema and his earlier removal by Edgar Lungu. Again, an appointment is not evidence of partisan military conduct. But the broader pattern matters when security institutions become increasingly prominent in a political contest.
Opposition activities faced police restrictions and security personnel became highly visible around politically sensitive events. After voting, the official residence of opposition presidential candidate Brian Mundubile was raided. Authorities alleged that shots had been fired at law-enforcement officers and that those arrested were connected to an alleged insurrection; the opposition disputed the government's account and described the operation as political repression.
There was also an allegation that officials raided opposition premises and confiscated money intended to pay polling agents. If confirmed, this would be particularly serious. Polling agents are part of the accountability architecture of an election. Their ability to observe voting and counting is one of the safeguards against manipulation. The seizure of funds intended to support their deployment would therefore have a direct electoral consequence and should be independently investigated.
**Election day: peaceful but not without problems **
It is important not to exaggerate what happened on election day. Voting was peaceful in most parts of the country. The Southern Africa Development Community (SADC) Electoral Observation Mission observers found the atmosphere calm and peaceful at 275 of the 277 polling stations they visited and, generally, found polling and counting orderly.
That matters.
But peaceful polling does not automatically equal a fully credible election. There were reports of delays in opening some polling stations, particularly in opposition strongholds. Cold weather also affected turnout in some areas, with some frustrated voters reportedly leaving without voting. The SADC raised concerns about the secrecy of the ballot, including the positioning of polling booths, as well as other procedural shortcomings. The secrecy of the vote is not a technicality. It is one of the protections against voter intimidation.
The accreditation process was also chaotic. Observer and stakeholder accreditation reportedly experienced delays and technical problems, with the ECZ moving between online and paper-based systems. This matters because elections are credible partly because people are able to watch them. When observers, journalists and civil-society organisations experience difficulty obtaining access, one layer of public verification is weakened.
**Then the counting stopped **
The most troubling moment came after polling. As results began to emerge, there were competing claims about the emerging outcome, including claims based on independent parallel voter tabulations that the incumbent was performing less strongly than expected. The claims require independent verification; they should not simply be treated as proof of who won.But the ECZ's subsequent decision to suspend counting and the announcement of results created a serious institutional problem.
The commission cited violence against election officials and the theft of election materials. Those are legitimate concerns if established.But suspending the counting process creates a break in the continuous chain between polling-station results and the national declaration. Once the chain is interrupted, the burden on the electoral commission to demonstrate what happened during the interruption becomes much greater. This is why the suspension was so damaging to public confidence. It created a black box at the most sensitive point in the election.
**Inside Mulungushi **
I was at the national results centre at Mulungushi International Conference Centre in Lusaka after completing my field observation. Having observed every presidential election since 2006, I can say that the atmosphere I encountered was markedly different from previous results centres I had experienced. The approach to Mulungushi was heavily securitised. It felt less like entering the nerve centre of a civilian democratic process and more like approaching a military installation. That matters because the national results centre should be the safest place in the country for political disagreement. Political parties should challenge the commission with documents; the commission should respond with evidence; observers should verify; journalists should report.
Instead, opposition representatives challenging aspects of the process were confronted by security personnel and removed from the immediate proceedings. I also witnessed the ECZ announce results which, in my assessment at the time, had not been subjected to the verification process that should precede national announcement. The observation does not prove electoral fraud. It does, however, reinforce the need for an independent audit.
**The Gen-20 problem **
The most serious allegations concern what happened to Gen-20 results forms between polling stations, district centres and the national results centre. There are allegations that some town clerks and other officials normally involved in the district process were replaced or sidelined; that intelligence officers became involved in collecting Gen-20 forms; that results were altered; that original forms with unfavourable results were destroyed or replaced; and that officials who resisted were intimidated.
The allegations are extraordinary and must be treated accordingly. They should not be accepted simply because they are made by the opposition. But neither should they be dismissed simply because the government denies them.
They are forensically testable.
That is why an independent audit is essential. The audit should compare original polling-station results with copies held by political parties, observers and the ECZ. It should examine alterations, overwriting, missing or duplicate forms and discrepancies between polling-station, constituency, district and national totals. It should reconstruct the chain of custody of the forms and establish who handled them, when and where.
If the official results reconcile with the polling-station evidence, the audit strengthens the ECZ and the government. If discrepancies exist but do not affect the outcome, that should be demonstrated. If discrepancies are sufficiently widespread to affect the result, Zambia must confront that evidence. The point is not to manufacture a different winner. It is to establish whether the declared winner won.
**Zambia has been here before **
There is an important lesson from 2021.The UPND was then the opposition. There were moments when it objected to the ECZ's handling of results and argued that the commission was moving towards announcing results before they had been sufficiently verified. Gary Nkombo and others stood firm and the commission ultimately changed course. The lesson is not that the UPND was necessarily more democratic in opposition than it is in government.The lesson is that political parties often understand the importance of electoral safeguards most clearly when they fear losing. That is precisely why institutions must be stronger than political parties. An electoral commission cannot be trusted only when it produces the result that one supports.
Zambia's democratic test
The preliminary assessments from SADC, the EU, the International Conference on the Great Lakes Region, the Electoral Support Network of Southern Africa, and AU-Comesa, alongside reporting by Amnesty International, Human Rights Watch and Zambian civil-society organisations, do not provide a simple answer to who won the election. They should not. But taken together, they paint a picture of a contest in which the political environment was increasingly restricted, the opposition faced structural and administrative difficulties, questions were raised about institutional independence, the electoral process became heavily securitised, accreditation was problematic, the secrecy of the ballot was contested and the results-management process ultimately generated profound distrust.
None of this proves that Hichilema did not win. But it does mean that the burden of proof has shifted. The government and the ECZ should not ask Zambians simply to trust them. They should show the evidence.
There is also a broader political paradox. Hichilema's government can legitimately point to improvements at the macroeconomic level, including progress on debt restructuring and renewed investor confidence. But macroeconomic improvement does not necessarily translate into improved household welfare. For many Zambians, the cost of food, electricity, transport and employment remains the more immediate measure of economic success.
Economic achievements deserve recognition. They do not purchase immunity from democratic accountability. Hichilema knows what it means to be an opposition politician. He spent years arguing that Zambia's institutions needed to be more democratic and elections more credible. In 2021, he benefited from the country's most important democratic achievement: an incumbent president lost and peacefully left office. That is the legacy he inherited.
The real test of his democratic credentials is therefore not whether he can win an election. It is whether he is willing to protect institutions strong enough to allow him to lose one. Zambia should not be asked to choose between political stability and electoral integrity. It can and must have both. Nor should this election be resolved through competing claims of victory, allegations of insurrection or the intimidation of political opponents.
The country has a simpler option.
Open the evidence. Audit the Gen-20 forms. Reconstruct the results chain. Let an independent process determine whether the national result can be traced back to the ballots cast at polling stations.
If Hichilema won fairly, the process will vindicate him. If there were irregularities that did not affect the result, quantify them. If the evidence demonstrates that the declared result cannot be reconciled with the votes cast, Zambia's institutions must have the courage to confront that reality.
That is what democratic accountability looks like. The danger facing Zambia is not necessarily that elections will disappear. It is something subtler: that elections continue to be held while the institutions that make them meaningful gradually lose their neutrality and credibility. Zambia has been one of the region's important democratic exceptions precisely because its political system has historically allowed power to change hands through the ballot box. It must not lose that distinction.
The ballots have been cast. Now the institutions must prove that the ballots matter.
Washington Katema is an international democracy, governance and human rights consultant with more than 20 years of experience. A long-time Zambia watcher, he has followed the country’s political and electoral developments for two decades and has observed every Zambian presidential election since 2006. He has undertaken elections related assignments in more than 30 countries, advising on electoral governance, political transitions and democratic accountability.