A month on: what went wrong with the Colombia referendum?

The future of the peace deal which was rejected by the Colombian people is still uncertain.

Sign Up

Get the New Statesman's Morning Call email.

The diplomatic calendar has witnessed an eventful past few months. There’s been the small matter of the US election. The US and China formally joined the Paris global climate agreement. Prime Ministers May and Trudeau delivered their debut addresses to the UN General Assembly. President Obama and Secretary-General Ban Ki-Moon attended for the last time. A few corridors along, following a failed ceasefire in Syria, the Security Council witnessed a walkout by the US, UK and French envoys in a heated exchange seeing both Syria and Russia accused of war crimes in Aleppo. 

Given these geopolitical developments, sparse attention has been given to one of the potentially more positive stories – the end of the last and the longest armed conflict in the Western hemisphere. A historic peace accord between the Colombia government and the FARC, ending over half a century of war which saw a quarter of a million killed and eight million displaced. The Colombian President, Juan Manuel Santos, was awarded the 2016 Nobel Peace Prize for his efforts to bring peace to the country. Nevertheless, in a surprising result, which echoes the current trend for shock political results around the world, the peace deal was rejected in last month's referendum, with 50.2 per cent of voters choosing to reject the deal, in a low 37 per cent turnout.  What went “wrong”? How did the voters choose to back the status quo over change, albeit by a vey narrow majority?

Here, conflict resolution specialists from Stratagem International shares their analysis, based on their work on the design stages of the Colombian referendum.

The campaigns were run like elections, but referendums are not elections  there were no candidates and no posts to win, and although parties were involved, they and the voters were seeking to influence an issue. Political campaigners therefore needed to “unlearn” their election instincts. While first mover advantage went initially to Yes (just as with the Good Friday Agreement in Northern Ireland two decades before, then the No campaign successfully “reframed” the terms of debate by questioning whether the negotiated peace deal could have been “better”. This was fatal – it put the positive proposition on the defensive, and also appropriated the Yes campaign’s peace rhetoric. The No campaign’s “We are not against peace. We are for a better peace” proved a brilliant message. It replayed the 2014 Presidential election themes, but Santos’s 52 per cent then was dented by the impact of fear triumphing over hope. Hurricane Matthew also intervened, hitting the country’s coastal regions at the start of October, deterring thousands of Caribbean region voters.

The voters answered the wrong question (as Charles de Gaulle groaned, after his final referendum loss) – as with many recent referendums, the Colombian outing was susceptible to “capture”. Voters used the plebiscite to register a protest against the government of the day, and against political leaders. Santos was polling disastrously, mid-term and with a stagnant economy – the voters were punishing him and his governing elite (even though Uribe’s No was also from that elite).

Emotion trumped reason; fear trumped hope — against a background of conflict, referendums often become more emotional than rational — voters express their instincts, rather than their cold, evidence-based selves. They remember the past and are reluctant to embrace an uncertain, or overly idealistic, future. Colombia was no different as the anti-amnesty purists caught the emotional zeitgeist, sowing fear of FARC leaders in power with impunity for terrorists. In other words, fear trumped hope.

Evidence shows that most referendums are lost (albeit narrowly) — promoting the “change” case is harder, especially if the agreements are complicated, lengthy, recent, and involve tough concessions — unless there is consensus that the change is overwhelmingly acceptable. The No campaign had these advantages – a 297-page agreement, a mid-term vote (when governments tend to be less popular) and harsh economic times (with a tough tax reform bill looming), making the risks of change seemed higher. The regions most affected by conflict voted Yes, but in many Colombian cities (excepting Bogota, Barranquilla and Cali), voters were already somewhat immunised from violence – why risk change?

It is easier to win a “No” campaign in a referendum — No supporters were able to scatter objections and complaints, with impunity, while the Yes side had to articulate its change proposition coherently, cogently and above all, consistently. Trying to explain away ten “free seats in Congress for FARC” without winning an election, for two terms, without going to prison, was never likely to be persuasive.

National polling fuelled voter complacency and apathy — referendums are more volatile and uncertain than elections — with last minute shifts in opinion. Most polls showed a clear advantage of 30 points in favour of Yes, and with every published poll since the agreement indicating endorsement, those soft Yes voters mistakenly thought it was a done deal. All polling failed spectacularly.

The No campaign conveyed consistency and unity — voters in referendums tend to look first to their political party of choice for advice, and then seek other cues from opinion-formers (churches, labour unions, NGOs, artists, celebrities and athletes). People especially like to see traditionally opposing politicians putting aside their differences in the national interest on platforms to promote their unified case, especially if this contrasts with the opponents. This did not happen for Yes, which had many diverse voices, most of them well coordinated; but the overly strong messaging emanating from an unpopular government undermined one of the key messages of the Yes campaign – that “this is not Santos’s peace, this is our peace.” It never became a “people to people” conversation; Uribe’s single No campaign was able to provide cohesive negative repetition. And it was fiercely fuelled by hatred of the FARC and the unpopularity of Santos.

Media reporting polarised opinions and played the election game — referendums are rarely well reported by the media, especially where there is no referendum culture (as, say, in Italy or Ireland). The Colombian media sought to portray a “gladiatorial” contest, polar opposite positions, argument and conflict, exactly as in the tightly fought 2014 Presidential election. However, the policy content of a plebiscite should permit textured discourse — shades of grey should be encouraged, not pummelled into submission — doubt and concern are legitimate feelings. The Santos vs. Uribe contest dominated the Colombian media.

Quintin Oliver and Ryan Gawn have recently returned from Bogotá where Stratagem International advised on the Colombian Referendum campaign. Oliver was Director of Northern Ireland’s Good Friday Agreement YES Campaign.