Oxford Q-Step Centre

Could Descriptive Statistics Help Facilitate Policy Convergence on UN Security Council Reform?

Perhaps no institution embodies the deficiencies of global governance as much as the United Nations Security Council (UNSC). It is the world’s sole legitimate authority for enforcing peace and security, and yet it often fails to execute its constitutional remit when its services are most in need. Rwanda (1994), Bosnia (1995), Darfur (2003), and Syria (from 2013) are but a few historical episodes etched in popular consciousness as emblematic of the institution’s inefficacy.

It is no wonder that serious calls to reform this beleaguered body date back more than two decades. Reform advocates have impugned the legitimacy of the UNSC on account these policy failures, and they have pointed to its antiquated membership design as the reason why; the UNSC has not changed since 1966, when the ranks of non-permanent members increased from six to ten. To redress the legitimacy gap, UN members created the Open-Ended Working Group (OEWG) in the mid-90s as a platform for tacking Security Council reform. Numerous policy proposals have surfaced since, some more feasible than others,[1] but the OEWG continues to plod along a seemingly interminable path.

Why have UN members failed to achieve reform? Most observers point to the policymaking process. Changes to the UN Charter require a two-thirds majority in the UN General Assembly (UNGA) and the consent of the permanent members (more on that below). But there might be more agreement among members than perhaps meets the eye. In this blog, I shed light on how reform proposals are premised on similar principles, yet struggle to identify areas of policy convergence because they decline to measure these principles in a systematic way. Descriptive statistics may offer a common language for reformers and in doing so reveal an expanded “win-set.”

For those new to UN politics, a few background details are worth mentioning here. (For those up to speed, skip to the next paragraph.) The UNSC is composed of fifteen member states. Five are permanent members (P5)—China, France, Russia, the UK, and the US—each of whom enjoys a constitutional entitlement to a seat at the table. There are a further ten seats belonging to non-permanent members (NPMs). NPMs are elected for two-year, non-recurring terms by the entire UNGA.[2] The major condition on these seats is that they follow a regional distribution: Africa (3); Asia-Pacific (2); Latin America and the Caribbean (2); Western Europe and Others[3] (2); and Eastern Europe (1). This means that the ten sitting NPMs always reflect this geographical allocation. Finally, each of the fifteen UNSC members has one vote, and nine affirmative votes are required for motions to pass. However, and crucially, each P5 member wields a veto on ‘substantive’ matters, such as investigations or Chapter VII enforcement actions. Therefore, every UNSC decision addressing a threat to peace and security must pass muster with the P5.

Reform proposals tend to coalesce around three areas: expansion and distribution of the membership, veto power, and working methods of the Council. Recommendations vary dramatically on how to tackle these areas. For instance, the Ezulwini Consensus (a reform plan designed by the African Union) calls for two veto-wielding permanent seats for African states whereas Uniting for Consensus (a reform plan designed by a coalition of twelve countries)[4] rejects any expansion to permanent membership. Yet, virtually all ideas for reform invoke the value-driven argument that the UNSC is unrepresentative and fails to reflect the current global distribution of power.

Consider the following comments made by key UN reform groups on the floor of the UNGA last November:[5]

  • Group of Four. ‘Here in New York, we painfully observe the shortcomings of the Security Council in its outdated composition, reflecting the year 1945, and its inadequate working methods.’[6]
  • Uniting for Consensus. ‘We owe [reform] to those who believe in a… modern Security Council that is more representative, democratic, accountable, transparent and effective.’[7]
  • African Union. ‘Africa’s position on Security Council reform is well known. Specifically, it is the only region that is not represented in the permanent category while being, at the same time, underrepresented in the Council’s non-permanent category.’[9]

These comments draw on common values. But what is so striking about the proposals that lie behind the excerpts is that they neglect to quantify the importance of these values; none of the proposals use descriptive statistics in any concerted way to advance their claims. Doing so not only would allow for comparability between policy options but also potentially facilitate policy convergence on UN Security Council reform.

To illustrate, let us examine one of the reform areas—expansion and distribution of the membership—using a few descriptive data. And to keep things simple, I focus on the category of non-permanent members. When the UNSC last expanded in 1966, relative parity existed across the five regions. The ratio of states (within each region) to seats (held by each region) ranged from 9.0 in Eastern Europe (EE) to 12.5 in Asia-Pacific—not identical but a decent approximation.

But regions have grown disproportionately since then, due to decolonization and the breakup of the former Soviet Union, leading to skews of representation over time (see Table 1). Today, the Western Europe and Others group (WEOG) holds the most favorable ratio at 13:1 while Asia-Pacific continues to hold the least favorable ratio at 26:1. In other words, WEOG gets an NPM for every thirteen states in its caucus whereas the Asia-Pacific group only gets an NPM for every twenty-six. It is shocking that such data are not cited by reform groups, especially those whose members suffer from underrepresentation.


To redress the inequity, a reformed NPM category would bring ratios back in line with each other by increasing the number of non-permanent seats. This step would involve allocating each additional NPM seat to the region that would most reduce skew. A simple way of measuring skew[10] is to take the absolute values of the differences between each regional ratio and the average of all ratios and then add them together. This gives us a single value that encapsulates overrepresented regions, e.g., Western Europe and Others, and underrepresented regions, e.g., Asia-Pacific.

I have done this in Table 2 below. Each row corresponds to a different number of NPM seats, ranging from ten (the current setup) to seventeen seats (a somewhat arbitrary stopping point). The rows include the number of seats allocated to each region, above which rest the members-to-seats ratios. The far-right column is my calculated skew value for each size of the non-permanent membership. As we can see, the skew values reduce dramatically when the NPM seats are expanded to 14 or more and distributed as stipulated in the table. In fact, each of the last four rows has less skew than did the UNSC in 1966 (7.2).


It is important to note that the principle of representation at work in these tables, as well as in leading reform proposals, is one of sovereign equality—the notion that every state is treated equally. Of course, we could use descriptive statistics to locate an equitable reform policy using an alternative principle, such as strength (which would employ a GDP measure) or individual equality (which would employ a population measure). And indeed, if we were to apply similar statistics to the permanent membership—with or without the veto—we would have to use a measure approximating power.

What this exercise helps illuminate is the idea that we can cut through the noise around UNSC reform by using some descriptive statistics to evaluate various models for equity and fairness. Existing reform proposals share similar values, but statistics may offer a common language to promote their harmonization.

     [1] See pages 38-39 of Richard Butler’s “Reform of the United Nations Security Council” (2012) for details on what the author concedes is a “naïve, wild, or even foolish proposal.”

     [2] Non-recurring means that non-permanent members cannot immediately run for reelection and must therefore wait at least one election cycle before standing again. There are no term limits.

     [3] “Others” is a UN term referring to industrialized states, including Australia, Canada, Japan, Korea, and New Zealand, among other countries.

     [4] These countries are Argentina, Canada, Colombia, Costa Rica, Italy, Malta, Mexico, Pakistan, San Marino, Korea, Spain, and Turkey.

     [5] UNGA, 71st Session, 42nd Plenary Meeting, 7 November 2016 (A/71/PV.42).

     [6] Harald Braun (Germany) on behalf of the Group of Four (Brazil, India, Japan, and Germany).

     [7] Sebastiano Cardi (Italy) on behalf of the Uniting for Consensus group.

[8] Menissa Rambally (Saint Lucia) on behalf of the L.69 Group of Developing Countries.


     [10] I am using the term in reference to bias toward a particular group rather than the asymmetry of a probability distribution.