Simulation Gaming OnlineFirst, published on April 14, 2010 as doi:10.

1177/1046878110365514

Simulation & Gaming

XX(X) 1­–30
Weighted Voting in the © The Author(s) 2010
Reprints and permission: http://www.
United Nations Security sagepub.com/journalsPermissions.nav
DOI: 10.1177/1046878110365514
Council: A Simulation http://sg.sagepub.com

Jonathan R. Strand1 and David P. Rapkin2

Abstract
Various observers have proposed weighted voting as part of an overhaul of the UN
Security Council.This article employs game-theoretic methods to simulate the effects
of weighting votes in the UN Security Council according to the ability of individual
states, as well as on the ability of the Council to act. The authors first briefly review
reform proposals. Drawing on recent proposals, they then calculate voting weights
based on population, contributions to the UN, and a measure of sovereign equality.
These component measures are assigned different weights in order to formulate
three sets of weighted votes. After this, they define several measures of voting power
that can be used to assess voting power in weighted voting systems. The measures
used in this article assess the relative voting power of individual members as well
as the probability the Council will engage in collective action to pass a resolution.
Results of voting power simulations are then presented under both simple and
qualified majority rules. The authors’ simulations indicate that under weighted voting
the capacity of the Council to act is likely to increase under most conditions unless
high qualified majorities are required.The simulations also demonstrate that the main
problem identified with weighted voting—large deviations between voting weight and
voting power—are not present in the weighted voting systems analyzed here.

Keywords
ability to act, capacity to act, deviations, high qualified majority, qualified majority rule,
reform proposals, simple majority rule, UNGA, United Nations General Assembly,
United Nations Security Council, UNSC, voting power, voting power indices, voting
weights, weighted voting

1
University of Nevada, Las Vegas, Las Vegas, NV, USA
2
University of Nebraska–Lincoln, Lincoln, NE, USA

Corresponding Author:
Jonathan R. Strand, Department of Political Science, University of Nevada–Las Vegas, 4505 Maryland
Parkway Box 455029, Las Vegas, NV 89154-5029, USA
Email: strand@unlv.nevada.edu

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

2 Simulation & Gaming XX(X)

The United Nations General Assembly (UNGA) and the United Nations Security
Council (UNSC) employ two different methods, based on different underlying
principles, for their members to make collective decisions. The UNGA’s one-country,
one-vote decision rule reflects the principle of the sovereign equality of states, as codi-
fied in the UN Charter (Chapter I, Article 2). The UNSC departs from this principle by
providing a veto to each of its five permanent members, in recognition of their much
greater power and influence. Each of these alternatives has serious shortcomings. In
the UNGA, Vanuatu, with a population of just over 200,000, has the same formal
influence as China, with a population of about 1.3 billion. In the UNSC, one perma-
nent member, say France with a population of 59 million, can veto initiatives favored
by all other UN members. The use of different decision rules in the two UN bodies
reflects a compromise aimed at reconciling the two contradictory principles. The codi-
fication of sovereign equality was contested by great powers who were understandably
concerned that members with smaller economies and populations would consistently
outvote them in the UNGA. At the same time, the vast majority of lesser powers
understandably did not want all decisions to rest with an oligopoly of great powers.1
Over time, the results of this compromise have been far from optimal: few matters of
consequence are decided in the UNGA, while numerous initiatives have been thwarted
by great power veto in the UNSC.
During its 60-year existence, the UNSC’s decision rules have been substantially
changed only once. In 1963, the size of the Council was increased from 11 to 15 mem-
bers in response to the accession of numerous newly independent states to the UN
system.2 As UN membership has increased by about 50% since that time, it is not
surprising that the current composition of the Security Council is often criticized for
not being representative of all UN members. The veto power afforded to the United
States, Russia, the United Kingdom, France, and China is regularly characterized as
reflecting the world system of 1945 rather than the 21st century. In 1993, the General
Assembly created the Open-Ended Working Group (OEWG) to examine possible
changes to the Security Council’s structure and decision-making procedures.3 The
OEWG focused on two “clusters” of possible changes (UNGA, 2004). Cluster I issues
include the size of the UNSC, increases in the number of permanent and nonperma-
nent members, and the veto power. Cluster II issues involve consideration of changes
that would improve the interaction between the Assembly and the Council, examina-
tion of ways to increase the participation of nonpermanent members in Council
deliberations, and ways to increase the UNSC’s transparency and accountability.4
The ongoing work of the committee and proposals from outside the UN has
focused attention on UN reform in general and Security Council reform in particular.
Calls for comprehensive reforms to improve the effectiveness and legitimacy of the
UNSC have become more numerous in the wake of the impasse between the Council
and the United States over the latter’s proposals and subsequent actions in Iraq during
late 2002 and early 2003. More general perceptions of growing U.S. resort to unilat-
eralism have added to these concerns. The rationale for reforming the Security
Council is summarized by Sutterlin (1996), who contends that

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 3

the effectiveness of the Security Council depends very heavily on the respect
in which it is held by member states . . . If the interests of the majority of
member states are not more adequately represented than is now the case, it is
unlikely they will, over the long run, comply with the Council’s decisions.
(pp. 9-10)

Furthermore, as Dervis (2005) argues, “radical reform is needed so that council

decisions be perceived to be much more legitimate than they are today” (p. 59). One
observer who backs the Dervis proposals noted that “it is absurd that Japan and
India—the world’s second biggest economy and its second most-populous one—have
less say in the Security Council than Russia, France or Britain” (Mallaby, 2007). Thus
far, the OEWG’s narrow mandate has not included consideration of weighted voting
as an alternative decision method. Weighted voting proposals have been floated by a
number of observers over the past 60 years. Some of these proposals have called for a
complete overhaul of UN decision-making procedures (e.g., Newcombe, 1983). Others
called for narrow and specific voting reform, such as the Kassebaum amendment—
first proposed in 1985 and again as recently as 2006—that would have required the
UNGA adopt weighted voting for making budgetary decisions.5 Schwartzberg (2003)
calls for substantial changes to the Charter in order to reallocate the distribution of
influence of states within the UN by means of weighted voting: Members’ votes
would be weighted according to national contributions to the UN, population, and
sovereign equality. Dervis (2005) offers a different weighted voting system in which
votes are weighted based on population, national product, contributions to global
public goods, and military capabilities. The appeal of weighted voting is, in part,
because the component measures are seemingly apolitical and perceived as fair to all
participants. Weighted voting systems are used in a variety of institutional settings,
notably in the International Monetary Fund (IMF) and World Bank (Rapkin & Strand,
2006; Strand & Rapkin, 2005). Perhaps the most closely studied weighted voting
system is the Electoral College in the United States.
This article addresses the question of how weighting votes would affect the rela-
tive influence of members of the Council, as well as the ability of the Council to act.
We begin with a brief overview of proposals for the weighting of votes from which
we define the voting weights we use in the simulations. The following section defines
several simple games that are used to analyze voting systems, especially voting sys-
tems employing weighted voting. By applying them to the hypothetical voting
weights, we are able to simulate the relative power of members of the Council under
weighted voting. Voting weights are not the best proxies for relative voting power of
members in weighted voting systems. Voting power is defined as a voter’s ability to
influence outcomes. Therefore, a particular voter’s voting power is a function of its
votes, the votes held by others, and the decision rule needed to pass a resolution. Our
simulations assess the weighted voting formulas with regard to the relative influence
of individual members of the Council. After this, we simulate how weighted voting
affects the abilities of members to block resolutions and initiate resolutions. Finally,

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

4 Simulation & Gaming XX(X)

we simulate how the different weighted voting schemes would affect the capacity of
the UN to act as a collective body.

Weighted Voting Schemes for

United Nations Security Council Reform
There have been several proposals for and assessments of weighted voting in the UNSC
and the UNGA (e.g., Dixon, 1983; Manno, 1966; Morrill, 1972; Newcombe, 1983;
Newcombe, Wert, & Newcombe, 1971; Newcombe, Young, & Sinaiko, 1977; Stassen,
1991). Tadokoro (1997), for example, proposes that the council adopt a weighted
voting system similar to that used by the IMF. While he does not provide specific
details on how to apportion votes, he suggests that votes “could be calculated from
such indicators as population, assessments, contribution of contingents to peacekeep-
ing, and voluntary financial contributions” (p. 128). Permanent members’ veto power
would be replaced by adopting higher majority requirements. But as observed half a
century ago by Rudzinski (1955), formulating

a system of weighted voting which would do justice to differences in size and

development, on the one hand, and on the other, prevent the crushing of the
smaller powers by the sheer weight of numbers inherent in a combination of
major powers represents a task of complexity and difficulty. (p. 370)

This task—balancing the interests of large and small states—requires the weighting
of votes based on some agreed-on set of component factors that measure national size
and power. In addition to assigning voting weights, Security Council reforms would
also have to specify the majority decision rule (e.g., simple majority, 60%, 70%) and
determine how members are to be selected for Council seats.
Recent proposals for reform of UN voting procedures have been made by Schwartz-
berg (2003, 2004) and Dervis (2005), who both maintain that the UN must reform its
decision-making processes in order for it to adapt to the geopolitical realities of the
21st century.6 In particular, the one-country, one-vote principle and the veto power
undermine the ability of the UN to operate effectively. Accordingly, reform must begin
with abandonment of the one-country, one-vote principle adhered to in the United
Nations. As Dervis (2005) argues, a weighted voting system should “reflect the actual
size, ability to act, and importance of the participating nation-states” (p. 59). These
proposals also call for the entire membership of the Security Council to be selected
recurrently from the overall membership of the UN.7 Concomitantly, de jure veto
power in the Council would also be abolished.
Sovereign equality as the underlying principle of vote allocation would be replaced
by a system of weighted votes. Recent proposals are based on three principles that
authors argue need to be reflected in the weighting of votes. First, there should be a
democratic and demographic element to weighted votes (Dervis, 2005; Schwartzberg,
2004).8 Measures of democracy, civil rights, or population are often proposed for

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 5

inclusion in weighted votes. A strong case can be made that the foremost international
organization, and one that purports to be democratic, should somehow take population
into account in the formal apportionment of influence.9
A second basis for weighting votes rests on the idea that the states that contribute
the bulk of the UN’s budget should have a greater say in the organization’s operations.
Schwartzberg (2003) rejects the use of direct measures of national product, advocating
instead the use of assessments, that is, paid contributions to the UN, a yardstick that
largely, but not entirely, derives from national product.10 Dervis (2005), on the other
hand, argues in favor of using both GDP and contributions to “global goods,” which
he measures as a state’s contribution to the UN’s regular budget. Assessments are not
a direct function of GDP or other measures of national product. The most important
divergence involves the United States: its contribution is capped by agreement at 22%,
but the U.S. share of global GDP is over 30%. Also, though assessments largely reflect
measures of national product, such as GDP, assessments for most developing coun-
tries are smaller than their shares of GDP, and assessments for developed countries are
larger. Thus, any shift toward a closer correlation between GDP and assessments
would shift votes toward developing countries. This shift would be even more favor-
able to developing countries if a measure of national product that adjusts for purchasing
power parity (PPP) were used instead of exchange rate measures.
Finally, there is often mention of some measure of sovereign equality. Schwartzberg
(2004) proposes use of a set proportion of the total weighted votes that would be
divided equally among all members, regardless of size or power. In the nomenclature
of the international financial institutions, he is advocating the use of “basic votes.”11
Thus, while weighted voting on the one hand reduces the prominence of sovereign
equality in UN decision rules, using it in the determination of weighted votes would
retain it, albeit in a significantly diminished form.12
Using the above proposals for guidance, we defined voting weights by additively
combining the following component measures: percentages of the total population of
UN members, paid regular contributions to the UN, and states’ unit shares, that is,
basic votes equivalents; each measure is assigned equal weight (33.3%).13
In our simulation, we develop a hypothetical Security Council in which members
are elected by the General Assembly and then partitioned into voting groups deter-
mined partly on a geographical and partly on an ideological basis. (See Appendix)14
For example, we use a “Central Europe” group consisting of Austria, Germany, and
Liechtenstein. Other voting groups include the Arab League, the Commonwealth of
Independent States, and a residual group of “non-bloc” members. This formulation of
voting groups extends the UN practice of using regionalism as an organizing principle
but adds to it the possibility that like-mindedness or other criteria may be useful for
purposes of organizing member states and aggregating their preferences.15 Moreover,
voting groups are used in many other international institutions, where they provide
selective representation within more general universal and regional organizations.16
Why should population, contributions, and basic votes (especially the latter) count
equally toward the apportionment of votes and influence? Voting weights could just as

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

6 Simulation & Gaming XX(X)

Table 1. Select Voting Weights Under VW Formulas

Country Contributions GDP Population VW1 VW2 VW3

United States 22.00 32.53 4.82 9.11 12.33 17.17

Japan 19.47 12.51 2.15 7.38 10.40 14.93
Germany 3.00 8.66 6.22 1.39 3.52 4.81
United Kingdom 11.00 6.13 4.90 1.00 2.55 3.44
France 17.00 6.03 4.48 1.00 2.52 3.39
China 2.05 3.97 21.49 8.02 6.53 4.29
Russia 1.80 1.08 2.45 1.59 1.64 1.72
India 0.42 1.60 17.44 6.13 4.70 2.56
Brazil 1.52 1.44 2.91 1.65 1.62 1.57
Nigeria 0.04 0.13 2.19 0.92 0.70 0.37
Note:VW = voting weights. Contributions are 2005 data; GDP and population are 2002 data.Values are
expressed in percentage of world share.

easily be apportioned with alternative, asymmetrical weightings for the component

factors. Furthermore, it is unlikely that the wealthier members who contribute most to
the UN budget—especially the present permanent members, who would also be giving
up this status as well as their veto—would agree to an arrangement in which basic
votes are as important as contributions. Instead, they would insist greater weight be
given to assessments. We therefore calculate three sets of voting weights derived from
different weights of each of the three measures.
Our first set of voting weights (VW1) use equal weights for population, contribu-
tions, and sovereign equality.17 The second set of weighted votes (VW2) is calculated
with contributions weighted at 50% and population and basic votes 25% each. The
third set (VW3) is calculated with contributions set at 75% and population and basic
votes at 12.5% each. In Table 1, we present component measures and the voting
weights for selected countries and include GDP for comparison purposes. These
three formulations of weighted voting produce very different distributional out-
comes. China, for example, has 8% of all votes in VW1, but this percentage falls to
6.5% and 4.3% in the two formulations that emphasize population and basic votes
less and financial contributions more heavily. India, Nigeria, and Brazil likewise
have more voting weight under the VW1 than VW2 and VW3. Not surprisingly, the
United States has much more voting weight using the formulas that emphasize con-
tributions (12.3% and 17.2%) than in the first formula (9.1%). Likewise, Japan has
more voting power under VW3. For very small members, voting weights would be
greater under the first formula than in the latter two. For example, Tuvalu (not
presented in Table 1) would have almost three times more voting weight under VW1
(.17%) than under VW3 (.065%).
Prior analyses of international organizations that use weighted voting show that the
actual influence a member has over electoral outcomes often deviates from the per-
centage of votes it holds (Dixon, 1983; Dreyer & Schotter, 1980; Leech, 2002b, 2002d;

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 7

Strand, 1999). This is due to the fact that voting weights fail to account for the overall
number of members that can form winning (and blocking) coalitions and how voting
weights are distributed among them. In other words, reference to voting weights alone
does not fully describe the influence of actors. To illustrate this point, consider a com-
mittee composed of three actors that uses weighted voting and makes decisions using
a simple majority. Two of the three members of the committee have two votes each,
and the third has one vote. The first two members each have 40% of all votes, while
the third has 20%, but it would be erroneous to infer from voting weights that the third
actor has half as much influence or power than the other two. Since it takes three votes
to pass a resolution there are three possible minimum winning coalitions.18 Each
member of the committee is part of the same number of potential winning coalitions.
Put differently, they each can contribute just as much voting power to the passing or
blocking of resolutions, so each has one third of the influence in the committee. For an
even more striking example, consider a committee in which one member has 99 votes,
the second 98 votes, and the third only 2 votes. A simple majority decision rule in this
committee requires 100 votes to pass an initiative. Even though the third member has
only a small fraction of the voting weight of the first two, it has an equal share of the
influence over electoral outcomes because of the manner in which minimum winning
coalitions can form. Recent proposals for weighted voting in the UN do not address
this potential pitfall of weighted voting systems (Dervis, 2005; Schwartzberg, 2003)
Another important aspect of voting systems is whether a simple or special majority
is required to pass an initiative. Currently, the Council employs a 60% majority
requirement for both procedural and substantive decisions, with the additional require-
ment that no substantive resolution passes if a permanent member votes against it. In
many international organizations, higher majority decision rules are common for sen-
sitive decisions. In the IMF, for example, an 85% majority is required, inter alia, for
constitutional reforms of its Articles of Agreement and for changes in the distribution
of quotas and votes. In order to discuss the impact of special majorities, we provide
analysis of four majority decision rules: simple, 60%, 70%, and 85%. The first two are
presently used in the UN itself; the latter two are employed in the IMF and World
Bank. Obviously, higher majorities make passing resolutions more difficult, but we
want to assess how the decision rule can affect the likelihood of agreement and
the impact on individual members’ abilities to affect outcomes. Before presenting the
results of the simulation, we first define several measures of voting power.

Voting Power and the Ability of the Collective to Act

In this section, we define measures of voting power. The first four assess the voting power
of individual actors in probabilistic voting games. A fifth measure assesses the probability
that a voting body will act (Coleman, 1971). These measures have been applied to a variety
of institutional contexts, such as the IMF (e.g., Leech, 2002d). The two most frequently
used measures of voting power are the Shapley and Shubik (1954) and the Banzhaf (1965)
indices. The indices defined by Coleman (1971) have not been applied as extensively.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

8 Simulation & Gaming XX(X)

In recent weighted voting proposals, only voting weights are considered not the
relative voting power of states in the UN (Dervis, 2005; Schwartzberg, 2003). As dis-
cussed above, consideration of voting weights does not account for the manner by
which coalitions may form. If the voting power of an actor can be defined as its ability
to influence outcomes, then a particular actor’s voting power is a function of its votes,
the votes held by other actors, and the decision rule needed to pass a resolution. This
is why measures of voting power, and not just voting weight, are needed to simulate
the effects of weighted voting in the UNSC.
A simple weighted voting game with n players as is conventionally defined as

[q: w1,w2,...,wn]
There are n players each with voting weight wi (i = 1,..., n) The number of votes
needed to pass a resolution is q. Coalition S is defined as a nonempty subset of the
players where

S ⊆ {1,2,...,n},

where S is winning coalition if and only if

Σ wi ≥ q
i∈S
Furthermore, we assume there cannot be disjoint winning coalitions. In other words,
n
q > –12 Σ wi.
i=1
The voting weight of S is expressed as

w(S) = Σ wi.
i∈S

In 1954, Shapely and Shubik published what is generally regarded as the first game-
theoretic article in the American Political Science Review. The measure of voting power
they defined became known as the Shapley-Shubik index (SSI). This measure assesses
actors based on their abilities to serve as “pivotal” (sometimes called swing) members of
winning coalitions. In short, a pivotal actor, by joining, turns a losing coalition into a
winning coalition. This ability to serve as the pivotal voter is considered a form of voting
power. SSI denotes the number of times (i.e., orderings) a player is pivotal divided by the
total number of possible orderings.19 Formally, the SSI for player i is defined as

Σ (n–S)!(S–1)!
SSIi = [v(S) – v(S – {i})],
n!
i∈S
S⊂N

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 9

where n is the number of players in the game, and s is the number of players in the
winning coalition, S. [v(S) – v(S –{i})] represents the weight added to coalition S as
pivotal player i becomes a member of the winning set of players.
The Banzhaf index (BPI) considers the number of winning coalitions where, for
player i in coalition S:

w(S) – wi < q ≤ w(S) + wi

Player i is considered critical to S, and S is considered a vulnerable winning coalition.

For a vulnerable coalition S, let V(i) denote the number of players critical to S. BPi is
the total number of times player i is a critical member:

BPi = Σ V(i).
i=1
This equation is an absolute measure generally referred to as the Banzhaf value. To
obtain a relative measure of voting power we normalize as follows:

BPi .
BPIi =
Σ n
k=1 BPk

A player’s Banzhaf voting power value is the number of times it is a critical player in
vulnerable winning coalitions divided by the total number of times all players are
critical in such coalitions. Values of the SSI and the normalized BPI sum to 1 and can
be interpreted as percentages.
Coleman (1971, 1986) defined three measures of voting power in response to
what he considered shortcomings of the SSI. The first two considered here are the
power of a player to initiate action (PTI) and the power of a player to block action
(PTB) in a voting system. (Note that among the three Coleman indices, PTI most
closely corresponds in meaning to the SSI and BPI.) In the context of the UNSC,
PTI means the capability to initiate a passing substantive or procedural resolution.
Estimating PTI and PTB entails identifying swing players. The PTI for player i is
defined as

ηi
PTIi = .
(2n–ϖ)

PTI for player i denotes the proportion of times it is a swing member to the number
of nonwinning coalitions.
Following Leech (2002a, 2002c), we define the number of times player i is the
swing player in the formation of a blocking coalition as hi. The power to block for
player i is defined as

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

10 Simulation & Gaming XX(X)

ηi
PTBi = ,
ϖ

where v is the number of winning coalitions. If player i can always block action by
the voting body, that is, it is a veto player, then its PTB is 1.0. Note that PTB and PTI
values are the same for player i if a simple majority decision rule is used. Also, unlike
BPI and SSI, PTB and PTI values cannot be interpreted as percentages because they
are absolute rather than relative measures of influence. For a given game, they range
from 0 to 1 but do not sum to 1.
Whereas the indices discussed so far generate estimates of each member’s voting
power, the third Coleman index estimates the capacity of a voting body to act, that is,
the probability that a collectivity will vote in favor of taking action. The power to act
(PTA) is simply the total number of winning coalitions divided by 2 raised to the nth
power where n is the number of actors:
ϖ
A= .
2n

And where v again denotes the total number of coalitions where w(S) ≥ q. The
denominator represents the total possible number of coalitions. The Coleman PTA is
best thought of as the probability that a voting body will pass a resolution—its
“decision probability.” When the required majority is increased, the number of possible
winning coalitions decreases; therefore, we expect higher decision rules to have lower
decision probabilities. The PTA is an important consideration in discussion of reforms
as any alteration of Security Council rules should increase rather than decrease the
probability of collective action. It would make little sense to institute reforms that
diminished the capacity of the UNSC to act. Hence, we employ Coleman’s PTA index
to evaluate the impact of reforms on the likelihood that the resulting UNSC would be
able to take action. This measure should be especially useful in assessing the impact
of special majorities and of alternative ways to weight votes on the likelihood of the
Council passing a resolution.
The next section reports voting power estimates generated by the SSI and BPI, fol-
lowed by results for the Coleman PTI and PTB, and then for the PTA index.

Simulating Weighted Voting in the UNSC

Relative Voting Power:The Shapley-Shubik and Banzhaf Indices

Previous studies on voting power in the UNSC have confirmed that the nonpermanent
members have substantially less voting power than the permanent members. For
example, O’Neill (1996) applies the SSI and finds that the permanent members each
hold 19.6% of the voting power while the nonpermanent members each hold merely
.019%.20 O’Neill (1996) dismisses the BPI because it assesses all mathematically

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 11

possible minimum winning coalitions regardless of the order in which the coalitions
form. He states that, “[k]nowing that most members will vote for a resolution should
increase our expectation that a further member will vote for it” (p. 221). Dixon (1983),
in his application of the Banzhaf voting power index to potential weighted voting
schemes (in the General Assembly), takes the opposite position—because the SSI
depends on the order in which coalitions form, it is “the least attractive”21 (p. 299).
This expectation, however, defeats the purpose of a priori measures of voting power.
In view of the fact that both the SSI and BPI approaches to voting power are con-
tested—generally and specifically in application to the UN—we report results for both
for purposes of comparing how they perform across the different voting weights and
majority requirements examined.
What then can we expect voting power indexes to tell us about how the distribu-
tion of power within the UNSC would be affected if weighted voting was implemented?
How are these results altered when we simulate changes to the weighted voting formu-
las? What effects do different majority requirements have on the relative power of
member states? To address these questions, we use the three sets of voting weights
defined above to analyze the distribution of voting power. The three tables discussed
in this section have nine columns of data.22 The first column presents the voting
weight for each member of the reformed Council. The other columns present voting
power results for the BPI and SSI under four different majority decision rules. Values
are reported in decimals, but since columns sum to 1, the data can be discussed in
percentage terms.
Table 2 displays results for the first set of voting weights. The voting group that
controls the most votes is the one in which Canada is the largest member; it holds
about 9.6% of all the votes. This number is very close to its voting power—9.9% and
10%, respectively, as measured by the BPI and SSI under a simple majority decision
rule. At higher majority thresholds, the two indices diverge: at the 85% level, the SSI
values for the larger voting groups in the Council are very close to their voting weights,
while their BPI values are significantly lower. The U.S. BPI value, for example, falls
by more than one fifth (.93 to .73) when we shift from a simple majority to an 85%
majority. The smaller voting groups gain BPI voting power, while their SSI values are
essentially unchanged. Overall, the simulation demonstrates that the distribution of
votes resulting from the weighed voting do not confer decisive voting power advan-
tages to any member or group of members.
The second set of voting power calculations estimates voting power when the
distribution of votes are determined by an alternative formula in which contributions
are weighted 50%, and population and basic votes are each weighted 25%. Using
this formula, the larger members, as expected, gain voting weight as compared with
the first set of voting weights: the United States has the highest voting weight (.122),
followed by Japan’s (.103) and Canada’s voting groups (.09). The SSI again proves
insensitive to changes in majority requirements, as all UNSC voting groups have
approximately the same SSI voting power across the 50%, 60%, 70%, and 85%
majority levels. The BPI, however, consistently estimates that the three largest

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

12 Simulation & Gaming XX(X)

Table 2. Voting Weight and Relative Voting Power With Formula 1

Simple Majority 60% Majority 70% Majority 85% Majority

Country/Voting Voting
Group (VG) Weight BPI SSI BPI SSI BPI SSI BPI SSI

VG18 (Canada) .096 .099 .100 .097 .100 .090 .099 .073 .093
United States .090 .093 .093 .091 .093 .087 .093 .073 .089
China .080 .081 .081 .080 .081 .078 .082 .072 .084
Japan .073 .073 .073 .073 .074 .073 .076 .069 .081
India .061 .061 .061 .061 .061 .060 .059 .063 .068
VG1 (Egypt) .056 .056 .056 .056 .055 .056 .055 .059 .060
VG2 (Ethiopia) .054 .054 .053 .054 .053 .054 .053 .055 .052
VG9 (Mexico) .053 .053 .053 .053 .053 .054 .052 .055 .051
VG16 (Nigeria) .053 .053 .053 .053 .053 .053 .052 .055 .051
VG14 (Indonesia) .048 .048 .048 .048 .048 .049 .048 .051 .046
VG13 (Italy) .048 .048 .048 .048 .048 .049 .047 .051 .046
VG11 (United .048 .048 .047 .048 .047 .049 .047 .051 .045
Kingdom)
VG12 (Brazil) .047 .047 .046 .047 .047 .048 .046 .050 .044
VG17 (France) .047 .046 .046 .047 .046 .048 .046 .050 .044
VG3 (Germany) .042 .041 .041 .042 .041 .043 .041 .046 .040
VG10 (Pakistan) .039 .039 .038 .038 .038 .040 .038 .044 .037
VG6 (Poland) .036 .036 .035 .036 .035 .038 .036 .044 .037
VG5 (Russia) .029 .027 .027 .028 .028 .031 .028 .040 .031
Note: BPI = Banzhaf index; SSI = Shapley-Shubik index.

voting groups lose voting power as the majority requirement is increased; this loss
is slight until we reach the 85% level, at which the United States, for example, loses
one third of its BPI voting power compared with the simple majority case (.122-.81).
For all other members, BPI values are essentially unchanged, except at the 85%,
level at which their voting power increases; this increase is not substantial for each
voting group individually but could be important in terms of their ability to form
winning coalitions (see Table 3).
Table 4 presents results of the simulation using the third formula for voting weights:
contributions are weighted 75%, and population and basic votes are weighted 12.5%
each. Note that the United States and Japan, with 17.6% and 15.3% of the voting
weight, respectively, would have de facto vetoes on decisions made with an 85%
majority rule. The difference between the largest vote holder and the smallest vote
holder is, as anticipated, the largest using this third formula. Again, there is much less
variation in SSI values across the four majority levels. BPI values decrease for the
United States and Japan as the majority requirement increases; at the 85% level, the
United States loses almost half of its BPI voting power (.192 to .098), and Japan’s falls
by 38% (.157 to .098). All others gain voting power; for the 11 smallest voting groups,
the gains range between 25% and 40%.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 13

Table 3. Voting Weight and Relative Voting Power With Formula 2

Simple Majority 60% Majority 70% Majority 85% Majority

Country/Voting Voting
Group Weight BPI SSI BPI SSI BPI SSI BPI SSI

United States .122 .130 .131 .125 .131 .110 .130 .081 .135
Japan .103 .106 .107 .104 .107 .099 .107 .078 .101
VG18 (Canada) .090 .091 .092 .091 .092 .089 .092 .077 .089
China .065 .064 .065 .065 .065 .065 .064 .067 .070
VG17 (France) .060 .060 .060 .060 .060 .061 .060 .063 .063
VG11 (United .059 .059 .059 .059 .059 .060 .059 .062 .058
Kingdom)
VG13 (Italy) .057 .056 .056 .057 .056 .058 .056 .061 .056
VG3 (Germany) .055 .054 .054 .055 .054 .056 .055 .060 .055
India .047 .046 .046 .046 .046 .048 .046 .052 .044
VG1 (Saudi Arabia) .046 .046 .045 .046 .045 .048 .045 .052 .043
VG9 (Mexico) .045 .044 .044 .045 .044 .047 .044 .051 .042
VG12 (Brazil) .043 .042 .042 .043 .042 .045 .042 .049 .040
VG2 (Ethiopia) .040 .040 .039 .040 .039 .042 .039 .047 .039
VG16 (Nigeria) .040 .039 .039 .040 .039 .041 .039 .047 .039
VG14 (Indonesia) .039 .038 .038 .039 .038 .041 .038 .047 .038
VG10 (Pakistan) .031 .030 .029 .030 .029 .032 .030 .038 .032
VG6 (Poland) .030 .029 .029 .030 .029 .031 .029 .037 .030
VG5 (Russia) .026 .025 .025 .026 .025 .027 .025 .033 .025
Note: BPI = Banzhaf index; SSI = Shapley-Shubik index.

The findings from the simulations thus far can be summarized as follows:

• Comparing indexes: BPI and SSI track voting weights rather closely.23 SSI
values are almost invariant across different majority requirements.
• Comparing majority requirements: High majority requirements make it easier
for one member (or a group) to block initiatives it opposes, but, since it also
facilitates others forming blocking coalitions, it becomes more difficult to pass
initiatives that it favors. Using BPI, this generalization is supported for the
larger members or voting groups. Groups with less voting weight would regis-
ter voting power gains, especially at the 85% level. In this application, SSI is of
little use for assessing the effects of different majority requirements because of
its lack of variation.
• Comparing weighted formulas: The equal weights (33.3% for each component
factor) in the first formulation are the most egalitarian with regard to the distri-
butional effects of weighted voting with different majority decision rules. As the
weight assigned to assessments/contributions increases (from one third to 50%
to 75%), and that accorded to population and basic votes correspondingly
decreases, larger contributors gain voting power.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

14 Simulation & Gaming XX(X)

Table 4. Voting Weight and Relative Voting Power With Formula 3

Simple Majority 60% Majority 70% Majority 85% Majority

Country/Voting Voting
Group Weight BPI SSI BPI SSI BPI SSI BPI SSI

United States .176 .192 .195 .187 .195 .159 .194 .098 .208
Japan .153 .157 .164 .160 .164 .150 .164 .098 .208
VG18 (Canada) .085 .083 .083 .083 .084 .088 .084 .090 .082
VG11 (United .079 .078 .078 .077 .078 .082 .077 .086 .075
Kingdom)
VG3 (Germany) .077 .076 .076 .076 .076 .080 .076 .086 .074
VG13 (Italy) .073 .071 .071 .071 .071 .075 .071 .083 .071
VG17 (France) .048 .047 .046 .047 .046 .050 .047 .061 .041
China .044 .042 .042 .043 .041 .045 .041 .055 .035
VG12 (Brazil) .039 .037 .036 .037 .036 .040 .036 .051 .031
VG9 (Mexico) .034 .033 .032 .033 .032 .035 .032 .045 .029
VG1 (Saudi Arabia) .033 .032 .031 .032 .031 .034 .031 .044 .028
India .026 .025 .025 .026 .024 .027 .025 .034 .020
VG14 (Indonesia) .026 .025 .024 .025 .024 .027 .025 .034 .020
VG5 (Russia) .023 .022 .021 .022 .021 .024 .021 .030 .017
VG2 (Ethiopia) .021 .020 .020 .021 .020 .022 .020 .028 .016
VG16 (Nigeria) .021 .020 .020 .021 .020 .022 .020 .028 .015
VG6 (Poland) .021 .020 .020 .020 .019 .022 .020 .027 .015
VG10 (Turkey) .019 .018 .017 .018 .017 .019 .017 .024 .014
Note: BPI = Banzhaf index; SSI = Shapley-Shubik index.

We now turn to results for the Coleman measures of the power of individual players
to block and to initiate action.

Absolute Voting Power: Coleman’s

Power to Initiate and Power to Block Indices
As defined above, PTI and PTB are related measures of control over voting outcomes.
Since the values generated by the two indices are inversely related, PTB and PTI are
especially useful to demonstrate the trade-off that players face between being able to
pass resolutions they favor and ensuring they can block resolutions they oppose. Both
measures are sensitive to the majority decision rule employed. As we will discuss in
this section, using higher decision rules can increase members’ PTB values but cor-
respondingly decrease their ability to initiate resolutions. If members of the UNSC
under weighted voting seek to minimize the likelihood of their worst-case policy out-
comes passing, they may very well insist on decision rules that maximize their
blocking power. But such a “minmax” strategy is not cost free because it results in
reduced PTI. Currently five members of the Council—the five veto holders—have

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 15

0.35

0.30

0.25

0.20
PTI

0.15

0.10

0.05

0.00
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 0.98
Threshold

18(Canada) US China Japan India 1(Egypt)

2(Ethiopia) 9(Mexico) 16(Nigeria) 14(Indonesia) 13(Italy) 11(UK)
12(Brazil) 17(France) 3(Germany) 10(Pakistan) 6(Poland) 5(Russia)

Figure 1. Power to initiate action under VW1

Note:VW1 = first set of voting weights.

PTB values of 1, denoting their veto player status. The nonpermanent members of the
Council presently hold a PTB value of .099 each.24
Operating under the current rules, the PTI for permanent members is .0266, and for
nonpermanent members it is .0026. The PTI using the first set of weighted votes is
displayed in Figure 1. Even using a simple majority decision rule, the highest PTI
value is about .32. At the 60% threshold, even the smallest vote holder has a PTI value
that is several times larger than under current rules. For the United States, its ability to
initiate action would be higher under this scenario than it is currently as long as the
decision rule remains under 72%. Put differently, the United States would have more
influence in forming winning coalitions than it does at present. PTI values quickly
converge to near zero as the decision rule is increased.
In Figure 2 (based on a 50%, 25%, 25% weighting), the PTI values again indicate
that all members of the Council would have more influence in initiating resolutions
than at present. The range between the PTI values of the higher and lower vote holders
is wider under this weighted voting formula.
When Figure 2 is compared with Figure 5, the trade-off between PTI and PTB is
quite apparent. If, for example, the United States insisted on a 70% majority level to

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

16 Simulation & Gaming XX(X)

0.40

0.35

0.30

0.25
PTI

0.20

0.15

0.10

0.05

0.00
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 0.98
Threshold

US Japan 18(Canada) China 17(France)

11(UK) 13(Italy) 3(Germany) India 1(Saudi Arabia)
9(Mexico) 12(Brazil) 2(Ethiopia) 16(Nigeria) 14(Indonesia)
10(Pakistan) 6(Poland) 5(Russia)

Figure 2. Power to initiate action under VW2

Note:VW2 = second set of voting weights.

provide it with ample blocking power, it would have a PTI value much smaller than if
a simple majority was used.
Figure 3 reports PTI values for a weighted voting formula that weights contribu-
tions much higher than population and basic votes (75%, 12.5%, 12.5%). Japan and
the United States have the highest values. In fact, their PTI values converge around the
70% threshold. When there are higher thresholds, the PTI of all members converge
toward zero. Moving from a simple majority to a 60% majority almost halves the PTI
of all members.
Figure 4 presents members’ power to block action under the first weighted voting
formula and continuous majority decision rules. As expected, each member of the
Council has greater power to block as the majority decision rule increases. The four
largest vote holders—the Canadian voting group, the United States, China, and Japan—
reach higher PTB values at lower thresholds than other members. Under a simple
majority rule, PTB values range from .09 to .32. This range widens as the threshold
increases to about the 85% threshold, beyond which PTB values for all members con-
verge on 1 as the threshold nears 100%. In short, the decision rule has an obvious and

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 17

0.50

0.40

0.30
PTI

0.20

0.10

0.00
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 0.98

Threshold

US Japan 18(Canada) China 17(France)

11(UK) 13(Italy) 3(Germany) India 1(Saudi Arabia)
9(Mexico) 12(Brazil) 2(Ethiopia) 16(Nigeria) 14(Indonesia)
10(Pakistan) 6(Poland) 5(Russia)

Figure 3. Power to initiate action under VW3

Note:VW3 = third set of voting weights.

significant influence on the ability of members to block action. Comparing the PTB
values in Figure 4 with the PTI values in Figure 1 reveals that there is an unmistakable
trade-off between being able to block proposals and being able to successfully initiate
proposals.
Figure 5 reports PTB values for votes based on weightings of 50% for assessments
and 25% for both population and basic votes. The three largest vote holders (United
States, Japan, and China) clearly have more blocking power than the other members.
In the case of an 85% majority requirement, these three have PTB values higher than
.9. At this high threshold, even the smallest voting groups have considerable power
to block UNSC action. If the Council were to require a two-thirds majority, the U.S
would have a PTB value higher than .75, while the Council seat with the lowest
voting weight would have a PTB value of only about .17. Members who want a
higher majority rule to ensure they could block resolutions to which they are strongly
opposed would pay the price of making it easier for initiatives they favor to be
blocked by other members.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

18 Simulation & Gaming XX(X)

1.00
0.90
0.80
0.70
0.60
PTB

0.50
0.40
0.30
0.20
0.10
0.00
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00
Threshold
18(Canada) US China Japan India
1(Egypt) 2(Ethiopia) 9(Mexico) 16(Nigeria) 14(Indonesia)
13(Italy) 11(UK) 12(Brazil) 17(France) 3(Germany)
10(Pakistan) 6(Poland) 5(Russia)

Figure 4. Power to block under VW1

Note:VW1 = first set of voting weights.

Under the third set of voting weights (75%, 12.5%, 12.5%), the power of the United
States and Japan to block action is much greater than that of the other Council mem-
bers (see Figure 6). These two large vote holders’ PTB values quickly converge on 1
as the threshold is raised. The smaller vote holders have significantly less power to
block at the lower majority decision rules.
Figures 4, 5, and 6 clearly indicate that the threshold chosen will make consider-
able difference to individual members’ abilities to block action by the Council. The
current veto holders in the Council would have less overall ability to block action
than they have under current rules. The other members would for the most part
have more than their current PTB value of .099. In other words, under weighted
voting the capacity to block action will be more evenly distributed across members
of the Council.
The trade-off between PTI and PTB is consistent across the three weighted
voting formulas. Moreover, the above analysis demonstrates how sensitive indi-
vidual PTI and PTB values are to majority decision rules. Using a weighted voting
scheme, the current permanent members would see a decline in their individual

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 19

1.00
0.90

0.80
0.70
0.60
PTB

0.50

0.40
0.30

0.20
0.10

0.00
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 1.00
Threshold
US Japan 18(Canada)
China 17(France) 11(UK)
13(Italy) 3(Germany) India
1(Saudi Arabia) 9(Mexico) 12(Brazil)
2(Ethiopia) 16(Nigeria) 14(Indonesia)
10(Pakistan) 6(Poland) 5(Russia)

Figure 5. Power to block under VW2

Note:VW3 = second set of voting weights.

abilities to obstruct decision making. Nonetheless, depending on the threshold

required for passage, the largest vote holders would retain considerable blocking
power. For most thresholds, and for each of the weighted voting formulas, indi-
vidual members of the Council would have more power to initiate action. The next
section, rather than considering the position of individual members of the UNSC,
analyzes the probability that the Council as a collective body will pass a resolution
under weighted voting.

The Power of the Collective to Act

Figure 7 reports the probabilities that the Council will be able to pass resolutions
for each of the weighted voting formulas and for all four majority decision rules.
Note that for the simple majority case the PTA is .5, and this probability drops
precipitously as the threshold is increased. Under current rules the PTA of the
Security Council is .0259. If we read PTA values as strict probabilities, then the
Council presently has less than a 3% chance of passing a resolution. At the 60%
majority decision rule, the PTA values for the Council are .212, .225, and .264,

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

20 Simulation & Gaming XX(X)

1.00
0.90

0.80

0.70
0.60

0.50
PTB

0.40

0.30
0.20

0.10

0.00
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90 0.95 0.97 0.99
Threshold
US Japan 18(Canada)
China 17(France) 11(UK)
13(Italy) 3(Germany) India
1(Saudi Arabia) 9(Mexico) 12(Brazil)
2(Ethiopia) 16(Nigeria) 14(Indonesia)
10(Pakistan) 6(Poland) 5(Russia)

Figure 6. Power to block under VW3

Note:VW3 = third set of voting weights.

respectively, for the three weighted voting formulas. Put differently, there is an 8
to 10 times higher likelihood under weighted voting that the Council will be able
to pass a resolution. PTA values drop to near zero at the 85% majority level.
Weighted voting can significantly increase the likelihood the Council will be able
to pass resolutions, but the PTA index is very sensitive to the majority decision
rule. Interestingly, the third weighted voting formula, which provides more voting
weight to larger contributors, would result in a slightly higher PTA value at most
majority levels. Put differently, a more unequal distribution of votes would facili-
tate collective action.

Conclusions
This article has simulated the voting power of members of the UNSC under three
different weighted voting schemes applying multiple measures of both relative and
absolute power. In addition, we have analyzed the capacity of the Security Council to
act. We first calculated voting weights and then offered three weighted voting

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 21

0.50

0.45

0.40

0.35

0.30
Probability

0.25

0.20

0.15

0.10

0.05

0.00
68

80

83

86

89

92

95

98
50

53

56

59

62

65

71

74

77
0.

0.

0.

0.

0.

0.
0.

0.

0.

0.

0.

0.

0.

0.

0.

0.

0.
Threshold

VW1 VW2 VW3

Figure 7. Decision probability

formulas. We then calculated several measures of voting power, including the power
of the Council to act, taking into account different majority decision rules. Overall
our results demonstrate that the selection of a majority decision rule is a critical factor
for both individual voting power and the probability that the Council will be able to
pass a resolution.
The relative voting power of Council members would differ slightly under the
three formulas. In general, there is little difference between voting weights and
voting power as defined by the BPI and SSI when using the formula where popula-
tion, contributions, and sovereign equality are equally weighted. For voting
weights based on higher weightings for contributions, some differences emerge. In
summary, the disparities between voting weight and voting power found other
weighted voting schemes are not systematically present. Overall, implementation
of this voting weight scheme for the UNSC would result in relative voting power
that is roughly equivalent to voting weights, although the more the weighted
voting formula emphasizes contributions, the greater the disparities between
power and weight.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

22 Simulation & Gaming XX(X)

The evaluation of the power of individual members to block action or to initi-

ate action unambiguously demonstrates the trade-off between these two types
of power. Moreover, the PTB and the PTI are very sensitive to the majority deci-
sion rule. The higher the threshold required, the easier it is for actors to block
action. Conversely, the higher the threshold, the more difficult it is for actors to
initiate action. This is true for each of the three weighted voting power formulas.
It is conceivable that several levels of special majorities might be employed in
any reform of UNSC voting rules that involves weighted voting, especially on
the most sensitive and important issues (e.g., such as use-of-force authorizations
under Chapter VII of the Charter). The implication of our analysis for such a
scenario is clear: the higher the majority decision rule, the greater the PTB for
all members.
Finally, our results for the decision probability of the Council being able to act
under weighted voting show that its PTA decreases as the required threshold is
increased. If a system of special majorities were adopted by the Council to accompany
weighted voting, it is possible that the likelihood of the Security Council passing a
resolution would be just as low—or lower—than it presently is, a result that would
defeat one of the goals of such a change in decision rules.
We have not addressed the political feasibility of adopting weighted voting.
Instead, we have provided a simulation of voting power under the type of system
several observers have recommended over the years. Our aim has been to explore
some of the logical possibilities for reforming UNSC decision procedures by
means of weighted voting, without consideration of the political prospects for
such reforms. That said, we readily acknowledge that the likelihood of a move to
weighted voting in the short to medium term is low. We agree with Dervis (2005)
that reform “efforts should not be deterred simply because it is probably not pos-
sible to find a perfect weighted scheme for Security Council votes that all members
would accept as optimal” (p. 60). The simulation of voting power presented here
demonstrates that a weighted voting system derived from a proposal such as
Schwartzberg (2003) or Dervis (2005) could serve as an effective mechanism to
apportion influence in the UNSC. Weighted voting, while giving more influence
to great powers, could also increase the probability that the Council will act col-
lectively to pass resolutions and address global problems. Any structural change
and change to voting rules will require a two-thirds majority in the General
Assembly and a 60% majority in the Security Council, with no veto holder in
opposition. Obviously, the primary limiting factor is obtaining the consent of all
veto holders. In addition, countries with smaller contributions and populations
may well resist any modification that results in their loss of influence. As the
results of our simulations indicate, the way votes are weighted and the use of spe-
cial majorities can affect the influence members wield and the ability of the
Council to act. Certainly devising and implementing a weighted voting formula
for the UN would be an intensely political process, subject to manipulation by
powerful members both within and outside of the formal negotiating process.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 23

Rudzinski’s (1955) observation that this would be a “complex and difficult task”
is indeed an understatement.

Appendix
Voting Weights and Voting Groups (in Percentages)

Country/Group Contributions GDP Population VW1 VW2 VW3

Voting Group 1: Algeria, 1.79 2.13 4.78 5.64 4.68 3.23

Bahrain, Djibouti,
Egypt, Iraq, Jordan,
Kuwait, Lebanon, Libya,
Mauritania, Morocco,
Oman, Qatar, Saudi
Arabia, Somalia, Sudan,
Syria, Tunisia, UAE,
Yemen
Voting Group 2: Angola, 0.08 0.29 5.28 5.41 4.08 2.08
Botswana, Burundi,
Comoros, Democratic
Republic of Congo,
Eritrea, Ethiopia, Kenya,
Lesotho, Madagascar,
Malawi, Mauritius,
Mozambique, Namibia,
Rwanda, Seychelles,
Swaziland, Tanzania,
Uganda, Zambia,
Zimbabwe
Voting Group 3: Austria, 9.55 6.89 1.53 4.21 5.54 7.54
Germany, Liechtenstein
China 2.05 3.97 21.49 8.02 6.53 4.29
Voting Group 5: Armenia, 1.86 1.28 3.66 2.88 2.62 2.24
Belarus, Georgia,
Moldova, Russia, Ukraine
Voting Group 6: Albania, 1.08 1.61 2.17 3.67 3.02 2.05
Bosnia, Bulgaria, Croatia,
Czech Republic,
Estonia, Hungary, Latvia,
Lithuania, Macedonia,
Poland, Romania,
Serbia, Slovak Republic,
Slovenia
India 0.42 1.60 17.44 6.13 4.70 2.56
Japan 19.47 12.51 2.15 7.38 10.40 14.94

(continued)

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

24 Simulation & Gaming XX(X)

Appendix (continued)

Country/Group Contributions GDP Population VW1 VW2 VW3

Voting Group 9: Antigua 2.09 2.45 2.69 5.39 4.59 3.33

and Barbuda, Bahamas,
Barbados, Belize,
Costa Rica, Dominica,
Dominican Republic,
El Salvador, Grenada,
Guatemala, Guyana,
Haiti, Honduras, Jamaica,
Mexico, Nicaragua,
Panama, St. Lucia, St.
Kitts and Nevis, St.
Vincent and Grenadines,
Suriname, Trinidad and
Tobago
Voting Group 10: 0.61 1.26 5.93 3.91 3.08 1.85
Afghanistan, Azerbaijan,
Iran, Kyrgyz Republic,
Maldives, Pakistan,
Tajikistan, Turkey,
Turkmenistan,
Uzbekistan
Voting Group 11: 9.44 7.61 1.47 4.85 5.99 7.72
Denmark, Finland,
Iceland, Ireland, Norway,
Sweden, the United
Kingdom
Voting Group 12: Argentina, 3.21 2.86 5.88 4.76 4.37 3.79
Bolivia, Brazil, Chile,
Colombia, Ecuador,
Paraguay, Peru, Uruguay,
Venezuela
Voting Group 13: Andorra, 8.43 6.57 2.02 4.87 5.76 7.09
Greece, Italy, Malta,
Monaco, Portugal, San
Marino, Spain
Voting Group 14: Brunei, 1.11 1.96 8.33 4.88 3.93 2.52
Cambodia, Indonesia,
Laos, Malaysia, Philippines,
Singapore, Sri Lanka,
Thailand,Vietnam
United States 22.00 32.53 4.82 9.11 12.33 17.17

(continued)

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 25

Appendix (continued)

Country/Group Contributions GDP Population VW1 VW2 VW3

Voting Group 16: Benin, 0.10 0.33 4.59 5.36 4.05 2.07
Burkina Faso, Cameroon,
Cape Verde, Central
African Republic, Chad,
Republic of Congo, Cote
d’Ivoire, Equatorial Guinea,
Gabon, Gambia, Ghana,
Guinea, Guinea-Bissau,
Liberia, Mali, Niger, Nigeria,
Sao Tome and Principe,
Senegal, Sierra Leone,Togo
Voting Group 17: Belgium, 10.06 7.47 1.57 4.74 6.07 8.07
France, Luxembourg, the
Netherlands, Switzerland
Voting Group 18: Australia, 7.40 6.80 7.07 9.66 9.09 8.24
Bangladesh, Bhutan,
Canada Cuba, Cyprus, Fiji,
Israel, Kazakhstan, Kiribati,
South Korea, Marshal
Islands, Micronesia,
Mongolia, Myanmar,
Nauru, Nepal, New
Zealand, Palau, Papua New
Guinea, Samoa, Solomon
Islands, South Africa,
Timor-Leste,Tonga,Tuvalu,
Vanuatu, Democratic
People’s Republic of Korea

Acknowledgments
For useful comments on prior drafts, we thank Mark Boyer, Courtney Smith, Alexander Thomp-
son, and two anonymous reviewers. For research assistance, we thank Patrick W. Scales.

Declaration of Conflicting Interests

The author(s) declared no conflicts of interest with respect to the authorship and/or publication
of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research and/or author-
ship of this article:

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

26 Simulation & Gaming XX(X)

Support for this research was provided by the Department of Political Science at the
University of Nevada Las Vegas.

Notes
1. These concerns were reflected in an early proposal to weight the votes in an international
assembly based on population, production, and trade (Sohn, 1944). Proposals for weighted
voting in the General Assembly surfaced again in the 1950s and 1970s (Newcombe et al.,
1971).
2. There have been other important changes in composition and operations of the Council,
perhaps most importantly the replacement of the Republic of China by the People’s Repub-
lic of China and the transfer of the Soviet seat to the Russian Federation.
3. The complete title of the working group is the “Open-Ended Working Group on the Ques-
tion of Equitable Representation On and Increase in the Membership of the Security Coun-
cil.” For an overview of the OEWG’s work, see UNGA (2004), Fassbender (1998; Chapter
9), and Daws (1997).
4. One of the issues it is researching is the question of increasing the transparency of the Coun-
cil, yet ironically the OEWG’s meetings and research are not in the public domain.
5. Under this proposal, votes would be weighted based on financial contributions to the UN
and a two-thirds majority would be needed to approve the budget. The then president of
the UN Association, Edward Luck, observed, “[g]etting two-thirds of the nations to agree
to weighted voting would be like trying to make the Pope a Marxist” (as cited in Sciolino,
1985, p. 22).
6. For his much earlier ideas on the subject, see Schwartzberg (1973), where he was respond-
ing to Morrill’s (1972) proposal for the adoption of weighted voting. In his 1973 com-
mentary, Schwartzberg is decidedly skeptical of the political prospects for the adoption of
Morrill’s (1972) proposal. In his 2003 proposal, Schwartzberg does not call for weighted
voting within the UNSC. In our adaptation of Schwartzberg (2003) and Dervis (2005), we
assume weighted voting on the council will be used in lieu of the veto power.
7. Dervis (2005) leaves open the possibility that some members will be permanent, at least
during a transition period.
8. For a brief but compelling perspective on why the UN or other international organizations
are highly unlikely to adopt democratic decision methods, see Russett (1997). For a more
detailed skeptical argument, see Dahl (2003). Another useful review of UN reform is found
in Archibugi (1993).
9. For a useful overview of demographic factors, see McNicoll (1999). Schwartzberg (2003)
avers that ideally his proposed system would assign a majority of the voting power to demo-
cratic states but acknowledges that this desideratum is problematic because it is “difficult to
draw a clear line between democratic and non-democratic regimes” (p. 94).
10. Assessments are not a direct function of GDP or other measures of national product. The
most important divergence involves the United States: its contribution is capped by agree-
ment at 22%, but the U.S. share of global GDP is more than 30%. Also, though assessments
largely reflect measures of national product, such as GDP, assessments for developing
countries are smaller than their shares of GDP, and assessments for developed countries

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 27

are larger. Thus, any shift toward a closer correlation between GDP and assessments would
shift votes toward developing countries. This shift would be even more favorable to devel-
oping countries if a measure of national product that adjusts for purchasing power parity
(PPP) were used instead of exchange rate measures.
11. For example, the IMF, World Bank, African Development Bank, and Inter-American
Development Bank all provide members “basic” votes in addition to the weighting of votes
on the basis of economic criteria.
12. Manno (1964) suggests a similar three-factor basis for the allocation of voting weights
but she recommends per capita contributions. Manno also suggests that different factors
could be weighted differently based on the type of decision being made. Dervis (2005)
adds to the above three a measure of military capabilities that he considers to be a way to
assess the ability of a member to contribute troops to peacekeeping and peace enforcement
missions. Given the way such missions are organized and that most peacekeepers are from
developing countries, we have doubts about including military capabilities in the weighting
of votes. Consider, for instance, that the United States contributes only a handful of person-
nel for peacekeeping missions, yet its voting weight would increase enormously were a
measure of military capability included. We would favor a direct measure of contributions
to UN peacekeeping.
13. For alternative variables and formulas for weighting UN votes, see Newcombe (1983),
who considers no less than 25 options, including contributions, population, square
root of population, cube root of population, and more elaborate formulas incorporating
basic votes.
14. Judging by the campaigns waged to gain nonpermanent seats in the UNSC (see Malone,
2000), the alignment of states into voting groups, on whatever grounds, can safely be
expected to be an intensely political process.
15. Newcombe (1983) suggests a similar type of selective representation, including the end of
permanent Security Council membership. She also considers the possibility that the veto
power could be retained for certain sensitive decisions and the possibility of using a special
majority vote to override a veto.
16. Selective representation in other international organizations, however, has proven to be
problematic when used in conjunction with weighted voting. For instance, in the Inter-
American Development Bank, there are voting groups in which the largest vote holder in
the group can elect itself as the group’s representative.
17. We obtained population data from the World Bank’s World Development Indicators. Popu-
lation data are from 2002, and the scale of UN assessments data are from 2005. For coun-
tries whose data are missing in the World Development Indicators, we estimated values
based on the most recent CIA Factbook.
18. There are four winning coalitions in the example (100, 99, 98, 2). The members can form
winning coalitions as 99:98, 99:2, 98:2, and 99:98:2. The last coalition, however, contains
a superfluous member, and hence the coalition is not a vulnerable coalition.
19. The SSI has been described by Felsenthal and Machover (2004) as “a voter’s expected rela-
tive share in a fixed prize available to the winning coalition, seen in the guise of a simple
TU [transferable utility] cooperative game” (p. 9). Power as a prize, P-power, is a relative

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

28 Simulation & Gaming XX(X)

measure of power in that it is the expected utility to an individual player for being a
member of the winning coalition. BPI measures a different aspect of voting power: power
over electoral outcomes, I-power (Felsenthal & Machover, 1998, 2004; Chapter 3). I-power
is closest to the common interpretation of power as influence in the literature.
20. One of O’Neill’s (1996) main contributions is his analysis of the concept of satisfaction. As
applied by O’Neill, a member’s satisfaction depends on “the probability that the Council
would pass a resolution the member wants passed or reject a resolution the member wants
rejected” (p. 225). His notion of satisfaction is somewhat similar to Coleman’s (1986) mea-
sures of individual voting power.
21. Dixon (1983) clearly favors the Banzhaf voting power index; though he defines Coleman’s
(1971) indices he does not apply them.
22. Results for SSI, BPI, and PTA used Bräuninger and König’s (2003) program, which they
have modified slightly for our application. PTB and PTI values were obtained using the
program written by Leech (2001).
23. Dubey and Shapley (1979) have demonstrated that there is no unproblematic way to calcu-
late one from the other, yet SSI and BPI often produce similar results. While this is gener-
ally true, as Straffin (1994) has established, there are instances where these two measures
of power diverge (as they do in this article at the 85% level).
24. The 1963 increase in membership of the Council increased the PTB for nonperma-
nent members from .088 to the present value. Prior to the 1963 reform, the PTI value
for permanent members was .0286 and for nonpermanent members it was .0025. See
Coleman (1986).

References
Archibugi, D. (1993). The reform of the UN and cosmopolitan democracy: A critical review.
Journal of Peace Research, 30, 301-315.
Banzhaf, J. F. (1965). Weighted voting doesn’t work: A mathematical analysis. Rutgers Law
Review, 19, 317-343.
Bräuninger, T., & König, T. (2003). Indices of Power IOP 2.0 [Computer software]. Konstanz,
Germany: Department of Politics and Management, University of Konstanz.
Coleman, J. S. (1971). Control of collectivities and the power of a collectivity to act. In
B. Lieberman (Ed.), Social choice (pp. 277-287) New York: Gordon & Breech.
Coleman, J. S. (1986). Individual interests and collective action: Selected essays. Cambridge,
UK: Cambridge University Press.
Dahl, R. (2003). Can international organizations be democratic? A skeptic’s view. In D. Held &
A. McGrew (Eds.), The global transformations reader (pp. 530-547). London: Polity Press.
Daws, S. (1997). Seeking seats, votes, and vetoes. The World Today, 53, 256-259.
Dervis, K. (with Ozaer, C.). (2005). A better globalization: Legitimacy, governance, and reform.
Washington, DC: Center for Global Development.
Dixon, W. J. (1983). The evaluation of weighted voting schemes for the United Nations General
Assembly. International Studies Quarterly, 27, 295-314.
Dreyer, J., & Schotter, A. (1980). Power relationships in the international monetary fund: The
consequences of quota changes. Review of Economics and Statistics, 62, 97-106.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

Strand and Rapkin 29

Dubey, P., & Shapley, L. S. (1979). Mathematical properties of the Banzhaf power index. Math-
ematics of Operations Research, 4, 99-131.
Fassbender, B. (1998). UN Security Council Reform and the right of veto: A constitutional
perspective. The Hague, Netherlands: Kluwer Law International.
Felsenthal, D. S., & Machover, M. (1998). The measurement of voting power: Theory and prac-
tice, problems and paradoxes. Cheltenham, UK: Edward Elgar.
Felsenthal, D. S., & Machover, M. (2004). A priori voting power: What is it all about? Political
Studies Review, 2, 1-23.
Leech, D. (2001). ipdirect [Computer software]. Coventry, UK: University of Warwick.
Retrieved February 19, 2010, from http://www.warwick.ac.uk/~ecaae/ipdirect.html
Leech, D. (2002a). Designing the voting system for the Council of the European Union. Public
Choice, 113, 437-464.
Leech, D. (2002b). An empirical comparison of the performance of classical power indices.
Political Studies, 50, 1-22.
Leech, D. (2002c). The use of Coleman’s power indices to inform the choice of voting rule with
reference to the IMF governing body and the EU Council of Ministers (Warwick Economic
Research Paper No. 645). Coventry, UK: University of Warwick.
Leech, D. (2002d). Voting power in the governance of the International Monetary Fund. Annu-
als of Operations Research, 109, 375-397.
Mallaby, S. (2007, May 5). Bound to fail. Newsweek International. (pp. 14-20)
Malone, D. M. (2000). Eyes on the prize: The quest for nonpermanent seats on the UN Security
Council. Global Governance, 6, 3-23.
Manno, C. S. (1964). Weighted voting in the United Nations General Assembly: A study of
feasibility and methods. Unpublished doctoral dissertation, School of International Service,
American University, Washington, DC.
Manno, C. S. (1966). Selective weighted voting in the UN General Assembly. International
Organization, 20, 37-62.
McNicoll, G. (1999). Population weights in the international order. Population and Develop-
ment Review, 25, 411-442.
Morrill, R. L. (1972). The geography of representation in the United Nations. The Professional
Geographer, 14, 297-301.
Newcombe, H. (1983). Design for a better world. Lanham, MA: University Press of America.
Newcombe, H., Wert, J., & Newcombe, A. (1971). Comparison of weighted voting formulas for
the United Nations. World Politics, 23, 452-492.
Newcombe, H., Young, C., & Sinaiko, E. (1977). Alternative pasts: A study of weighted at the
United Nations. International Organization, 31, 579-586.
O’Neill, B. (1996). Power and satisfaction in the United Nations Security Council. Journal of
Conflict Resolution, 40, 219-237.
Rapkin, D. P., & Strand, J. R. (2006). Reforming the IMF’s weighted voting system. The World
Economy, 29, 305-324.
Rudzinski, A. (1955). Majority rule vs. great power agreement in the United Nations. Interna-
tional Organization, 9, 366-375.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016

30 Simulation & Gaming XX(X)

Russett, B. (Ed.). (1997). Ten balances for weighing UN reform proposals. In The once and
future Security Council (pp. 13-28). New York: St. Martin’s Press.
Schwartzberg, J. E. (1973). More on “the Geography of Representation in the United Nations”:
An alternative to the Morrill Proposal. The Professional Geographer 15, 205-210.
Schwartzberg, J. E. (2003). Entitlement quotients as a vehicle for United Nations Reform.
Global Governance, 9, 81-114.
Schwartzberg, J. E. (2004). Revitalizing the United Nations: Reform through weighted voting.
New York: Institute for Global Policy.
Sciolino, E. (1985, October 5). Kassebaum bill has UN worried. New York Times, p. 22.
Shapley, L. S., & Shubik, M. (1954). A method for evaluation the distribution of power in a
committee system. American Political Science Review, 48, 787-792.
Sohn, L. B. (1944). Weighted voting in an International Assembly. American Political Science
Review, 38, 1192-1203.
Stassen, H. (1991). We the peoples of the world. In F. Barnaby (Ed.), Building a more demo-
cratic United Nations (pp. 36-45). London: Frank Cass.
Straffin P. D. (1994). Power and stability in politics. In R. J. Aumann & S. Hart (Eds.), Hand-
book of game theory (Vol. 2, pp. 1127-1151). Amsterdam: Elsevier.
Strand, J. R. (1999). State power in a multilateral context: Voting strength in the Asian Develop-
ment Bank. International Interactions, 25, 53-74.
Strand, J. R., & Rapkin, D. P. (2005). Voting power implications of a double majority voting
procedure in the IMF’s executive board. In A. Buira (Ed.), Reforming the governance of the
IMF and World Bank (pp. 235-250). London: Anthem Press.
Sutterlin James S. (1997). The past as prologue. In B. Russett, (Ed.) The Once and Future
Security Council. (pp. 1-12). New York: St. Martin’s.
Tadokoro, M. (1997). A Japanese view on restructuring the Security Council. In B. Russett
(Ed.), The once and future Security Council (pp. 119-134). New York: St. Martin’s.
United Nations General Assembly. (2004). Report of the open-ended working group on the
question of equitable representation on and increase in the membership of the Security
Council and other matters related to the Security Council General Assembly (Fifty-Eighth
Session Supplement No. 47 [A/58/47]). New York: United Nations.

Bios
Jonathan R. Strand is an associate professor of political science and Program Chair for Asian
Studies at the University of Nevada Las Vegas. His research has appeared in World Develop-
ment, Journal of East Asian Studies, International Interactions, and elsewhere. His book, The
Regional Development Banks: Development Lending with a Regional Flavor, is forthcoming.
Contact: strand@unlv.nevada.edu.

David P. Rapkin is an associate professor of political science at the University of Nebraska. He

was a visiting professor at Tsukuba University in Japan during 1988-1990 and 1996-1998. His
research and teaching interests are in global political economy, with an emphasis on East Asia.
His articles have appeared in New Political Economy, World Development, The World Econ-
omy, and Review of International Studies. Contact: drapkin2@unl.edu.

Downloaded from sag.sagepub.com at PENNSYLVANIA STATE UNIV on September 16, 2016