Making Peace in Colombia: Is a Northern Irish Style Power-Sharing Possible?

Making Peace in Colombia: Is a Northern Irish Style Power-Sharing Possible?

By Luis Fernando Medina (Juan March Institute) and Neophytos Loizides (Senior Lecturer at the University of Kent).

To contact the authors, email:

Colombia’s peace process has reached an important milestone. Achieving peace seems more possible at the present stage but the process is likely to fail without an agreement on the most fundamental issue of sharing power. Other post-conflict societies such as Northern Ireland, South Africa and Bosnia faced comparable challenges in their post-conflict transitions but succeeded in ending violence despite conditions that initially seemed prohibitive. Specifically, Northern Ireland has the potential to inspire inclusive political arrangements in Colombia as the province has developed power-sharing institutions that have contributed to an unprecedented political stability.

Having agreed on a set of points on agrarian and land policy, the Colombian government and the FARC have now moved on to discuss the next point of the agenda, effective participation of the FARC in legal politics. This issue will be particularly thorny, especially given that it has been one of the main targets of the critics of the process. Although at first glance they seem are rather disconnected, there are some important connections between implementing the agreed land reforms and power-sharing; these connections pose important challenges not only for the initial stages of negotiating a peace settlement but also for the post-conflict implementation stage. In that sense, we believe that there are international precedents and lessons of constitutional design to benefit Colombia, particularly if adapted to the country’s local context.

Although the details are not yet known, it is possible to make some well-educated guesses about the guidelines of the different land policy issues already agreed. Some of these points have already been discussed openly with broad input from several sectors of Colombia’s civil society and it stands to reason that the agreed upon document reflects the major concerns that have been expressed during that stage. It is very likely that the end-product of the agreement will contain at least three elements:

  1. An effort to update the land registry so as to modernize rural tax assessments, discourage speculation with idle lands and clarify property rights.
  2. Some degree of recognition for existing peasant communities that will, through limits on the extension of land estates in some areas, dampen the operation of market forces that in the past have led to massive processes of internal migration.
  3. An ambitious program of investment in public goods in some of the most neglected areas of the agrarian periphery.

There may be more elements, and the precise combination of these three still remains an open question. But what is already clear is that these elements have one thing in common: they require a sustained effort at implementation that can last several years.

This prospect raises eerie historical parallels. Twice throughout the 20th century Colombia launched efforts at land reform, with bold legislative initiatives announced to great fanfare only to see them diluted and ultimately reversed few years later. There is a strong case to be made that the missed opportunities of these two reversals came back to haunt Colombia through renewed violence a few years later. So, it would be wise to use this third chance to address the issues of Colombia’s countryside in a more resolute way that can help the political settlement remain in place.

Here is where the questions of political accommodation come into play. Much has been made in the public debate about the participation of the FARC in politics, especially focusing on its possible arrival to Congress. For the most vociferous critics of the peace process, any participation of the FARC in politics is absolutely unconscionable. Some of those critics will never accept a settlement (and let’s hope they keep their reticence at the level of peaceful political expressions) but others are, doubtless, giving expression to a legitimate concern among Colombian society: the fact that some elements of the FARC (as in Northern Ireland) are guilty of grievous war crimes and violations of human rights. But let’s, for the sake of argument, assume that there are ways to address this issue satisfactorily. For example, participation in politics could be restricted to elements of the FARC not directly linked to crimes; some members of Colombia’s civil society, untainted with the war, could emerge as political spokespeople and so on. Such formulae have been tried in the past and in Northern Ireland where parties hitherto opposing each other’s participation in power-sharing have now emerged as successful partners in the executive. The question that needs to be asked then is whether democratic politics is the right locus of political participation in face of the daunting issues facing a post-conflict society.

In this regard, the experience of the peace agreement in Northern Ireland offers an interesting innovation that may be well worth considering: the d’Hondt executive. The idea is, simply, to create a mechanism that guarantees automatic participation of all the different political groups in the cabinet, with a share proportional to their electoral results.

Unlike majoritarian political systems, the d’Hondt executive does not operate as a winner-takes-all mechanism; it gives a share in decision-making and responsibility to minorities. Furthermore, it allows such minorities to retain their political identity as such since it allows them to enter governing coalitions after the electoral stage. The d’Hondt executive is fluid and democratic; no political party is assured a share in power unless it obtains electoral support at the polls; the better they do that, the more they will be rewarded with a larger share. In other words, it forces political parties to focus exclusive on politics as the only game in town and incentivizes to stay away from violence as the example of Northern Ireland suggests.

Figure 1: Hypothetical Illustration of the d’Hondt Allocation (from votes to cabinet seats)


Divide a party’s percentage in the polls by


Divide a party’s percentage in the polls by


Divide a party’s percentage in the polls by


Party A 34, 28%  (1st choice) 17, 14% (3rd choice) 11,42% (6th choice)
Party B 32, 67 % (2nd choice) 16, 33% (4th choice) 10, 89% (7th choice)
Party C 15, 76% (5th choice) 7,88% 4, 73%
Party D 8,93% (8th choice) 4,47% 2, 97%


Parties’ percentage in votes is divided by 1, 2, 3….(d’Hondt divisor)

Parties choose in order of highest quotient for eight positions in the cabinet (or more)

Order of picking ministerial positions among the four parties is A,B,A,B,C,A,B,D

The idea is not entirely novel in the Colombian context. The National Front was a ‘tacit consociation’ that gave the two major parties equal shares in the cabinet but not according to their democratic mandate, thus lacking long-term legitimation. Also, in the early 1990s the Gaviria Administration rewarded the strong showing of the demobilized guerrilla M-19 at the polls by giving it the Ministry of Health. The D’Hondt executive would be, in essence, a formalization of Gaviria’s precedent, one that would not go all the way to adopt the non-proportional formula of the National Front. The d’Hondt executive in Northern Ireland has proven to be a mechanism that gives incentives for bridge-building across political groups, overcoming situations of extreme polarization.

Conceivably, such a formula could bypass the problem already identified because it could give the defenders of the new land reform a seat at the table where the crucial decisions are taken, where the ministry and the relevant agencies are staffed, where the plans are designed, where the budgets are computed. Of course, they would need to obtain enough electoral strength and within the cabinet majority decisions will be made with reference to implementing the peace agreement. This will allow all kinds of positive synergies to develop. There is no doubt that the only way in which the FARC can fully re-enter the political system is if they can build bridges with other sectors of society. The d’Hondt executive would provide the incentives to do so and a clear system of rewards if they succeed.

There are, to be sure, critiques to be addressed before presenting this proposal as an element for debate. The most obvious complication is that it may require a move away from the presidentialism that has so marked Colombia’s political system. A fully operational d’Hondt executive would make the cabinet dependent on the legislature in ways that are novel to most Colombians but nonetheless negotiable as political parties might prefer a system that entitles them participation in government. Equally, it is important to remember that many scholars regard parliamentary systems and proportional representation as more durable and effective especially in dealing with post-conflict societies. And there is no need to go all the way to a parliamentary system, but certainly it would place the country along the path to a semi-parliamentary regime, akin to, say, France. Such a change has been debated in Colombia for years so it is not something that comes entirely out of the blue. This will allow Colombia to benefit from both systems but there is no denying that it would be a change that needs to be studied carefully.

For instance, the president could be primarily in charge of breaking deadlocks should they arise. In other words, he or she could be an arbiter among political parties rather than an additional veto player in preventing necessary reforms. Also, any nominee to occupy a portfolio in the cabinet could be first approved by the other parties as in Switzerland. To ensure appropriate monitoring, committees in parliament will he also allocated through the d’Hondt but headed by a representative from a different party from the one holding the relevant ministry in the executive.  Another option is to limit the operation of the system in time, as during the transition in South Africa or in Colombia itself in the past. Finally, d’Hondt could be agreed as a mechanism of portfolio assignments for three or four presidential terms, for instance and with a review process to follow in due course.

Colombia will need those alternative options and novel ideas in negotiating a future peace agreement; given the nature of the post-conflict situation, such options will be needed in the context of a sustained effort directed not only at legislating changes in the next few months but also implementing and sustaining them through for several years. Inevitably, the peace process will require profound constitutional changes, some of which seem initially at odds with Colombia’s current political system but could be highly beneficial for all sides in the long-term.