Printer Friendly

Institutional Design and Power Relations in the African Development Bank.

JONATHAN R. STRAND [*]

ABSTRACT

Unlike other international financial institutions (IFI), when the African Development Bank (ADB) was established it did not offer membership to advanced industrial states. In need of more capital, the ADB decided to admit these countries in 1982. In doing so, it attempted to institute rules to maintain the controlling influence of the regional, African members. One means to this end was to grant regional members (as a bloc) the largest share of votes. This paper analyzes how the addition of the non-regional members affected formal voting power relations. Voting power is calculated using the Banzhaf and Johnston power indicies over five time-points since 1975. The results indicate formal influence in the Bank has changed drastically since the 1982 restructuring. Voting power has been concentrated in a subset of members due in large part to the rise in the number of dictators in the Bank's voting groups. Some members that donate substantial capital have little or no voting power. The results point to a well-kno wn property of weighted voting systems. Namely, the intended appropriation of influence in the form of voting weights often fails to translate into relative voting strength in practice because voting weights need to be considered in relation to other factors such as voting procedures and decision rules.

Authors concerned with the role of international institutions in world politics have largely focused on questions of state agency, system structure, and whether or to what extent states are embedded in a web of institutions that restricts their behavior. Unfortunately, these authors often ignore the formal institutional rules by which decision-making, at least in theory, occurs. In the international financial institutions (IFI), governance takes places under explicit sets of binding rules. Key to any discussion of any international institution is the question of what rules members use to organize and operate the institution. Voting procedures in international institutions can be structured in a variety of ways. The United Nations General Assembly, for example, utilizes the logic of one nation, one vote. Other international assemblies explicitly grant some members more influence (e.g., the UN Security Council). Through the use of weighted voting, the IFI also base formal relations on the principle that some me mbers should have more influence.

Weighted voting is a common way formal influence is apportioned in WI. Members that contribute large amounts of capital want to maintain control over how the capital is managed. Hence, the percentage share of votes a country has in an IFI is roughly commensurate to its percentage share of capital contributions. In the regional development banks an additional consideration in the determination of voting shares is whether a country is a regional or non-regional member. Many studies of the IFI refer to voting weights -- the percentage share of votes -- as a proxy for the amount of influence held by members. [1] Voting weights, however, do not adequately measure a member's ability to affect outcomes because they do not account for either the possible number of coalitions that may form or the number of votes needed to pass a resolution. Measuring influence in a weighted voting system necessitates the use of simple games, commonly called power indices.

The case of the African Development Bank (ADB) is especially interesting from the perspective of formal power relations among members. Unlike the International Monetary Fund, the World Bank, and the other regional development banks, the ADB was formed without the direct influence of non-African states. It was not until 1982 that non-African members were admitted. The addition of non-regional members was problematic for regional members because it was feared the non-regional members could come to dominate the institution. Keep in mind the Inter-American Development Bank (IDB) is clearly dominated by the United States and that the United States and Japan jointly manage the Asian Development Bank (AsDB). [2] The prospect of non-regional members controlling decision-making in the ADB was anathema to regional members. There was, however, a clear need for an influx of capital and, ultimately, regional members devised a reformation of institution rules to allow the participation of non-regional members while simulta neously trying to preserve the African character of the Bank. The purpose of this paper is to examine formal power relations in the Bank before and after the 1982 institutional adjustment and assess the impact this adjustment had on power relations among members.

The influence of an actor in a voting system can be assessed in many ways. One approach is to calculate the percentage of total votes an actor holds. This is an especially appealing way to look at weighted voting systems where members have different numbers of votes. But the percentage of votes an actor holds does not necessarily describe its influence in the formation of winning coalitions. This is because voting weight -- the percent shares of votes an actor holds -- does not take into account the various ways actors can form coalitions. [3] The size of a minimum winning coalition is also a function of the type of majority rule needed to pass a resolution. If the voting power of an actor can be defined as its ability to influence the formation of winning coalitions, then a particular actor's voting power is a function of its voting weight, the voting weights of other actors, and the decision rule needed to pass a resolution. Therefore, reference to an actor's voting weight does not fully describe its influe nce over outcomes. A measure of relative voting strength that incorporates these other factors is needed.

This paper first briefly describes the institutional context within which decision-making occurs. The focus is on the formal rules by which the ADB is organized. The paper then turns to a discussion of why a measure of relative voting power is needed. Two measures of voting power are defined -- the Banzhaf and Johnston power indices. The Banzhaf index evaluates an actor's criticalness in the formation of coalitions. The Johnston index focuses on how uniquely critical an actor is relative all other critical actors. In the data analysis section, results for the Bank's Board of Directors are presented at five time-points starting in 1975. Power relations have been greatly altered by the admission of the non-regional members. Some members have little or no voting power because of selective representation on the Board of Directors. The conclusion focuses on how, despite the aims of its framers, the ADB's formal institutional rules have not always result in the intended distribution of influence. The use of weight ed voting in the ADB and other IFI must therefore be called into question.

Decision-making and Institutional Adjustment in the African Development Bank

The ADB was established in 1964 as a purely African institution with contributions coming only from African governments. Yet the ADB, in need of capital, agreed to allow non-regional countries membership in the early 1980s. Nonregional members entered in 1982, and for the first time non-African donors where given direct influence in ADB decision-making. [4] An attempt was made to abate the impact of the non-regional members by granting regional members the majority of votes. Unlike the AsDB where the United States and Japan have the most votes or the IDB where the United States is dominant, in the ADB "it is impossible for any specific country to control voting or exercise veto power" Mingst (1990:16). Compared to other IFI, the ADB's weighted voting system in considered egalitarian (Mingst 1990). But voting weights -- as will be shown in the next section -- do not measure relative influence over outcomes. In other words, reference to the number of votes held by members of the ADB falls to describe the diffe rent ways members can enter into collations. What matters is not the absolute number of votes a member has, but the number of votes a member has relative to all others.

The ADB's decision to alter its membership by admitting non-regional members presents one of the most distinct cases of institutional adjustment by a multilateral development bank. [5] In order to maintain a regional appearance post-1982, two thirds of votes are vested in regional members. There are two assemblies charged with decision-making authority: the Board of Governors and the Board of Directors. The Board of Governors is comprised of high-ranking finance ministers. The Board of Directors is in effect comprised of elected representatives from the larger Board of Governors. Initially there were nine Executive Directors on the Board of Directors. These nine representatives received the nine largest shares of votes in the Board of Governors. [6] Soon after the admission of non-regional members the 1990s the membership of the Board of Director's rose to 18. As in the case in the other IFI, day-to-day decision-making is delegated from the Board of Governors to the smaller Board of Directors. It is commonpl ace for the Board of Governors to merely approve resolutions, loans, and other actions that have been negotiated by the Directors. The relatively large size of the Board of Governors and the fact that it meets only once a year (while the Directors meet in continuous session) limits it as a decision-making body.

In order to sustain regional influence, the majority of Executive Directors must be from African members. All members of the ADB, except for the United States, are pooled into "voting groups." Each voting group elects -- using a simple-majority method -- an Executive Director to represent the group in the Board or Directors. Voting within the Board of Directors utilizes a simple-majority count for most items, including the passage of loans. [7] This form of selective representation, as the results presented in the next section reveal, often limits the influence of members with smaller voting weights.

While the basic principle of weighted voting in the IFI favors members that contribute larger amounts of capital, an effort is made to furnish smaller (usually borrowing) members with a voice in decision-making. All IFI make use of "basic" votes whereby each member is granted a minimum number of votes. The ADB provides members with 625 base votes and one vote for each share of capital held.8 The use of basic votes can be viewed as an attempt to balance the interests of the large members with the idea that all members should be able to participate in decision-making.

In sum, several institutional features are relevant for the analysis in this paper. First, the ADB employs a system of weighted voting. Attention is paid in the assignment of voting weights to whether or not a member is a "regional" or a "non-regional" member. This paper poses the question of how the addition of the non-regional members in 1982 impacted the voting power of regional members. Second, day-to-day decision-making is delegated from the Board of Governors to the Board of Directors. For this reason the paper focuses on voting power for the Executive Directors since they essentially manage the policy decisions of the Bank. The use of selective representation raises concern for the ability of small members to impact the election of a representative. Third, because of this delegation of authority, voting in the Board of Directors basically occurs in two stages. In determining representation on the Board of Directors, members of each voting group vote for an Executive Director. In the second stage the E xecutive Directors vote on resolutions. Finally, at each level a simple majority is needed to form a winning coalition. The potential impact these factors may have on voting strength suggests that reference to voting weights does not fully describe how coalitions can form. In other words, voting procedures may attenuate a large voting weight. Small weights may not necessarily denote little influence over outcomes. Voting weights do not account for the process by which representatives are elected to the Board of Directors and therefore do not assess influence in the more consequential decision-making organ of the ADB.

Voting weights have not been static. Figure 1 displays voting weights from 1969 to 1996 for several major shareholders in the ADB. Figure 1 displays how even the largest 9 members hold less than 40 percent of the votes. Indeed, until the entry of the non-regional members the 4 largest members held less than 30 percent of all votes. Nigeria has consistently held about 10 to 13 percent of the voting weight. Immediately after the entry of the non-regional members, Nigeria experienced a decline to about 8 percent. Algeria, Libya, and Zaire have all experienced decreases in voting weight and this decrease appears to be related to the 1982 institutional adjustment. After 1982, Algeria's voting power decreased from about 8 percent to 3.5 percent. It has since rebounded to the 4 percent range.

Nigeria was the largest shareholder during the entire time period. The United States became the second largest shareholder with 5 to 6 percent of the voting weight in the 1980s. Japan has remained in the 4 to 5 percent range, while Germany, France, and Canada each hold 3 to 4 percent. Since 1989 voting weights have remained roughly the same. In the IDB, the largest six members hold about 80 percent of the voting weight. The seven largest members in the AsDB hold 50 percent of its votes. In other words, from an analysis of voting weights, the ADB appears to have a more egalitarian distribution of votes. Voting weights, however, do not denote voting power. Moreover, how the addition of the non-regional members affected relative voting power remains to be addressed. The next section explains why voting weights cannot be relied on for assessing relative influence, and defines two measures of voting power.

Weighted Voting and the Measurement of Voting Power

Power, to say the least, is a pivotal concept in political science, yet it remains a contested concept with a myriad of definitions and interpretations (Ash 1951; Baldwin 1989). Power can suggest influence, domination, or control of one actor by another (Strange 1996:17-21). The use of the term power in this study has a specific connotation: the ability to influence outcomes. Early studies of power within committee systems attempted to quantify this definition of power (e.g., Shapley and Shubik 1954). [9] These studies helped establish a class of simple games commonly called voting-power indices. Power indices assess the potential of actors to affect the formation of all possible winning coalitions in a specific mathematically defined model.

Voting-power indices offer preference-free measures of power. As others have noted, preferences are important in the analysis of how actors interact in institutional settings (Garrett and Tsebelis 1996). Certain actors may be more likely to form coalitions based on mutual interests. The position taken here concurs with Brams (1985) in that "a complete definition of power must include both preferences and outcomes... that is, power is the causation of outcomes by preferences" (p. 107, note 6). Therefore, the use of a power index is best viewed as filling a supporting role for discussions of states' interests and capabilities. Preferences of actors in large international institutions are not readily known. The results obtained from power indices supplement our understanding of the decision-making processes of these institutions by supplying the relative strength of members -- on one dimension of power -- before preferences are expressed during negotiations. [10] It is important to emphasize that voting power va lues assess influence before any bargaining takes place and before any affinities based on preferences are considered. It is in this sense that they are considered a priori.

A simple voting game with n players is usually denoted:

[q; [w.sub.1], [w.sub.2],..., [w.sub.n]]

where there are n players each with voting weight w. The total voting weight required for passage is q. A coalition is any non-empty subset of the players. For any coalition S [subset or equal to] {1, 2, ... n}, the voting weight of S is

w(S) = [[sigma].sub.i[epsilon]S] [w.sub.i].

A coalition is wining if and only if w (S) [greater than or equal to] q. [11]

The Banzhaf and Johnston indices require calculation of vulnerable coalitions. A vulnerable coalition, S, is a winning coalition such that, for player i, w(S) - [w.sub.i] [less than] q [greater than or equal to] w(S). In this case i is considered critical to S. For vulnerable coalition S, let c(S) denote the number of players critical to S. The primary difference between the Banzhaf and Johnston indices is how each scores a player's criticalness. The Banzhaf index relies on the absolute number of times that members are critical. The Banzhaf index assigns a value of 1 to a player each time he or she is critical. The Johnston index identifies how uniquely critical an actor is in the formation of coalitions. It does so by splitting the 1 point among all critical actors in each coalition. A critical member receives 1/n points where n is the number of actors critical to the coalition. [12] For any player i, let V(i) = {S : S is vulnerable and i is critical for S}.

Banzhaf power ([BP.sub.i]) is the total number of times a player is critical. In other words:

[BP.sub.i] = [[sigma].sub.S[epsilon]V(i)] c(S)

Since the Johnston index differs in how it assigns values to critical members, the Johnston power for player i ([JP.sub.i]) is:

[JP.sub.i] = [[sigma].sub.S[epsilon]V(i)] 1/c(S)

Standardizing these BP and JP values, the Banzhaf (Bi) and Johnston (Ji) power indices, respectively, are defined:

[B.sub.wi] = [BP.sub.wi]/[[[sigma].sup.n].sub.k=1] [BP.sub.wi]

[J.sub.wi] = [JP.sub.wi]/[[[sigma].sup.n].sub.k=1][JP.sub.wi]

A player's Banzhaf power-value is simply the number of times that the player is a critical actor divided by the total number of times that all players are critical actors. Similarly, an actor's Johnston power value is its JP value divided by the sum of all players' JP values. Values for both fall between 0 and 1 and the sum of all values is 1; hence the results can be read as percentages.

Because the ADB employs two stages of voting, the overall voting power of an actor is the product of two intermediary values. Each value is the result of related, yet temporally separate, voting games. As Nurmi (1997) explains:

These games are nested in the sense that each player of a large 'macro' game is composed of several players playing a 'micro' game. The outcomes reached in the micro games determine the strategies of the players of the macro game and thus indirectly also the outcome of the macro game. (P. 321)

In order to obtain each member's voting power it is necessary to first calculate voting power within the voting groups. The voting power for the voting groups is then calculated. The product of each value is the overall voting power. For example, say a member has half of the voting power for a voting group that holds 10 percent of the voting power in the Board of Directors. That member's voting power is 5 percent of the voting power in the Board of Directors. This method to derive voting power is essentially the approach defined by Owen (1977; 1981) and applied by Hosli (1997) to assess the voting power of a priori coalitions. [13] Owen (1977) describes a method for obtaining voting power based on the probability that a particular coalition will form. The method used here sets this probability at 1 since the coalitions are not a matter of likelihood but are dictated by the institutional context.

Why is a measure of relative voting power is needed? Consider the following hypothetical voting game:

[100; 99, 98, 2]

One player has 99 votes, one has 98 votes, and one player has 2 votes. One hundred votes are required to pass a resolution (a simple majority). The player with 2 votes clearly has a disadvantage in terms of voting weight. In terms of how coalitions can form, however, each player is equally critical because each is a member of the same number of minimum winning coalitions. Any two players can form a winning coalition. If we only considered voting weights we would conclude that one player has little influence. But, in terms of how many times each player can influence outcomes, then, each player has equal influence. Therefore, in some circumstances voting weights are not very good indicators of relative influence.

Brains (1990) notes that, compared to the Banzhaf index, the Johnston index "distinguishes between being critical in a winning coalition when one is uniquely so and when one shares this critical role with other actors, in which case one's power is proportionately reduced" (p. 230). This distinction between the two indices can be demonstrated by considering the following example:

[70; 40,15,5,5,5,5,5,5,5,5,5,5]

In this situation there are 12 actors, one with 40 votes, one with 15 votes, and 10 with 5 votes. The Banzhaf value for each of the three weights is .381, .164, and .046, respectively; while the Johnston value for each is .626, .188, and .019. In the above example the Johnston index captures the fact that the player with 40 votes is a critical member for all winning coalitions. In other words, the Johnston index identifies this player's veto power. This result is similar to studies that have looked at the power of the U.S. president in passing (and blocking) legislation (e.g., Brains et al. 1989; Brains 1978). The Banzhaf index assigns the president about 4 percent of the voting power, compared to the Senate's 33 percent and the House's 63 percent. The Johnston index, in contrast, assigns the president 77 percent of the voting power, the Senate 16 percent, and the House only 7 percent. With the distinction between the Banzhaf and Johnston indices noted, the next section applies these measures to the ADB's Boa rd of Directors at five time-points since 1975.

Voting Power in the African Development Bank

The tables discussed in this section have seven columns of voting strength data. For the most part results are presented only for members that hold at least 1 percent of voting power on the Johnston index. Voting weights are listed in the first data column. Recall that voting occurs at two levels, and that it is the product of voting power at each level that determines overall voting power in the Board of Directors. [B.sub.m] and [J.sub.m] are the members' power values within the voting groups. [B.sub.s] and [J.sub.s] are the Banzhaf and the Johnston power values, respectively, for the voting groups. [B.sub.t] and [J.sub.t] are the "total" power values and are probably the columns of greatest interest. All values are derived using a simple majority rule. B, is a measure of criticalness while [J.sub.t] is a measure of unique criticalness.

Table 1 displays voting strength in the ADB for 1975. [14] Both Nigeria and Algeria were dictators in their voting groups. Dictators are able to guarantee their election to the Board of Directors. Nigeria, it turns out, is a dictator throughout this period because it is a dyadic voting group with the much smaller Sao Tome and Principe. Nigeria held the greatest voting weight in 1975 (9.6 percent), and this weight translated into a Banzhaf value of 13 percent, and a Johnston value of 16 percent. Algeria, in large part due to its dictator status, obtains more voting power than its 7 percent voting weight suggests. Interestingly, in one syndicate, Libya is only 178 votes away from being a dictator. Yet because the other 2 members of its voting group (Egypt and Upper Volta) are equally critical-actors, the three members share the group's voting power. Hence, Upper Volta, with only 1.3 percent of the voting weight has as much voting power as Libya (6.9 percent of voting weight). The first ten members listed in the table hold 47 percent of the voting weight, 57 percent of voting power on the Banzhaf index, and 66 percent of the voting power on the Johnston index. Dictators capture about 8 percent of all votes.

Table two presents results for 1980. [15] Libya joins Nigeria and Algeria as a dictator. Nigeria has roughly the same Banzhaf value as its voting weight and a Johnston value that is slightly higher than its voting weight. The top ten members hold 51 percent of the voting weight, 68 percent of the voting power on the Banzhaf index, and 73 percent of the voting power on the Johnston index. Overall the results for 1980 do not differ dramatically from 1975. For instance, voting weight and power is concentrated slight more in the largest members, and there is one more dictator. Within other syndicates, Egypt and Morocco capture the majority of the voting power. Yet as in 1975, within most syndicates voting power is shared across members; meaning that most members of the ADB have a voice in the election of representatives to the Board of Directors.

In 1985, after the admission of non-regional members, power relations among members of the ADB were altered considerably (see Table 3). Surprisingly, Japan has the highest level of unique criticalness as measured by the Johnston index, 21 percent. Japan dominates the largest voting group, holding over 86 percent of the voting power within the voting group on both power indices. Japan had less than 5 percent of the voting weight, but since it dominated the largest voting group it obtained surprisingly high voting power in the Board of Directors. The total numbers of dictators increased from 3 to 11. The increase in the number of dictators suggests many more members have little or no formal influence in the election of members of the Board of Directors. These members include several capital donating countries such as the U.K., Italy, Belgium, and the Netherlands.

The top ten members listed in Table 3 hold 44 percent of the voting weight, 60 percent of the voting power on the Banzhaf index, and 70-percent of the voting power on the Johnston index. This result is consistent with 1980 and 1975 except for the fact 5 non-regional members comprise the top ten. Nigeria has more voting power on the Johnston index than their voting weight suggests, but its Banzhaf value is only slightly higher than its voting weight. After the admission of non-regional members, Nigeria gained voting power on the Johnston index while losing voting power on the Banzhaf index (and losing voting weight). All other regional members listed in Table 2 lost voting power after the admission of the non-regional members. Indeed, some countries, such as Algeria, lost quite a bit of influence. Among the non-regional members, Germany, France, and the United States each hold about 5.5 to 6 percent of voting power on the Banzhaf index and 4.2 to 5 percent of voting power on the Johnston index.

Japan's dominance of its voting group evolved into outright dictatorship by 1990. Yet it was no longer in the largest voting group. Moreover, the United States, with individual representation, came within about 500 votes of the voting weight of Japan's voting group. Hence, even though Japan gained influence within its voting group and still had about the same voting weight, overall Japan's voting power fell dramatically from its 1985 value. By 1990 the number of dictators rose to 12. As Tables 4 and 5 reveal, the member that gained the most in the 1990s was Nigeria. [16] Nigeria's voting weight increased by about 2 percent from its 1985 value, and its Banzhaf value increased by about the same percent. What is striking is the increase in Nigeria's Johnston power-values, up to 35 percent in 1990 and then down to 29 percent in 1995. The next highest Johnston value in 1990 was Egypt's 5.7 percent. In 1990, the top ten members held 49 percent of the voting weight, 62 percent of voting power on the Banzhaf index, and 74 percent of the voting power on the Johnston index. The percentages in 1995 were 47, 60, and 69, respectively.

Thus far the focus has been on results for individual members. There is a limit of 33 percent on the total voting weight of non-regional counties. In 1985, due in large part to Japan's elevated voting power, the non-regional members held 38.4 percent of the voting power on the Banzhaf index and 45 percent of the voting power on the Johnston index. It can be said then that the non-regional members obtained more voting power than their voting weight suggests (and more than ADB institutional rules dictate). In 1990 and 1995 the non-regional members held 26 percent of the voting power according to the Banzhaf index and 35 percent according to the Johnston index. Hence on one index non-regional members have less influence than their voting weights are designed to confer and on the other they have about the same voting power as voting weight. But when voting power of individual members is considered, the admission of the non-regional members has definitely diminished the influence of most African members with the e xception of Nigeria. The concluding section considers implications of these results.

Conclusion

A central question of this paper is how do voting weights in the ADB translate into ability to influence outcomes? Due to the unique nature of weighted voting systems, measures of a priori voting power were required to address this question. The evidence presented above suggests that the post-1982 institutional design of the ADB grants some members little voting power while others are granted more power than their voting weights suggest. Dictators in the voting groups have greater influence over electoral outcomes than their voting weights imply. The majority of members in the ADB have no formal influence over electoral outcomes in the Board of Directors. Several members that contribute capital have no formal voting power in the Board of Directors due in large part to the nature of the first stage of voting whereby one member of a voting group dominates the election of a representative to the Board of Directors. Currently, 12 out of 18 voting groups have dictators, including 4 of the 5 non-regional voting gro ups. [17]

Dictators did not always dominate the Bank's voting groups. Before the admission of non-regional members only 2 of 9 syndicates had dictators and one of these voting groups had only 2 members. Exactly why there is a prevalence of dictators after the institutional adjustment is not clear. One possible way to reduce the number of dictators in the syndicates would be to increase the decision-rule from a simple majority to a special majority of, for example, 85 percent. This would afford members with 15 percent of the voting weight in their syndicates to influence the election of a representative. Raising the decision-rule, however, would furnish some members with de facto veto power; a prospect that may not be very appealing to other members. This result points to a well-known and fundamental property of weighted voting schemes: relative influence in weighted voting systems is often very different than the intended appropriation of influence in the form of voting weights. Some members dominate a voting group and thereby "capture" others' votes. According to one observer, "the United States lacks the structural power in the form of votes and veto ability granted to it by other development banks" but the United States still attempts to impose its will on the ADB (Shaw 1991:538). [18] If the United States wishes to increase its voting power without increasing its financial contributions it could try to form a voting group that it can dominate and consequently capture the other members' votes. [19]

Unlike most other African members who lost voting weight and power after 1982, Nigeria thrived. It is an even more critical actor in 1995 than in 1975. Nigeria's voting weight has not varied as greatly as its voting power, as measured by the Johnston index. By 1995 Nigeria had 6 times the voting power on this index than the next largest member. Having one member with significantly greater voting power than all others does not seem to fit with the framers' goals of an egalitarian distribution of influence. Moreover, the 625 base votes furnished to each member do not guarantee each member will have formal influence in the Board of Directors. Using measures of a priori voting power reveals that voting weights in the ADB do not necessarily equal voting strength. The manner by which voting occurs, the weighting of votes, and the type of majority used results in some members have no formal voting power.

The usefulness of analyzing voting power is to demonstrate how institutional design affects relative power sharing among members. While it is not the suggestion here that "weighted voting does not work," future alterations to institutional rules may wish to account for differences in voting power; not just voting weights. [20] Weighted voting may not be the most efficacious way to structure voting in institutions like the ADB. Weighted voting has been utilized by the IFI as a way to ensure that the states that supply the majority of the resources control lending and other activities. The United States and other developed countries view weighted voting systems as essential to maintaining their influence in the IFI. Indeed, American presidential administrations often refer to the influence that U.S. voting weight provides as a way to convince Congress to approve capital increases. Weighted voting may be an effective way to mollify these types of domestic issues. However, anyone who proposes weighted voting as a solution to the inequality of actors must first recognize that weighted voting is potentially a problematic technique to apportion influence.

(*.) Department of Political Science, Niagara University, NY 14109, U.S.A.

NOTES

(1.) For example, Gianaris (1990) only considers voting weights.

(2.) On voting power in the AsDB see Strand (1999). For an assessment of voting power in the IMF see Leech (1998).

(3.) As Schotter (1981) asserts, "[v]oting weights are extremely bad proxies for voting power" (p. 324). Power indices have been found useful in a variety of substantive settings, including assessments of the power held by voters in U.S. presidential elections (Lambert 1988; Eisner 1993), the weighting of votes in the United Nations General Assembly (Dixon 1983), the UN Security Council (O'Neill 1996), financial decision-making (Gambarelli 1994), and assessment of the decision rules involved in amending Canada's Constitution (Kilgour 1983; 1985; Levesque and Moore 1984). Applications to the international financial institutions have been limited. Dreyer and Schotter (1980) applied the Banzhaf index to changes in membership and voting weights in the IMF's Board of Governors before and after the 1978 quota review. Their results demonstrate what they term the "paradox of redistribution," whereby a member's voting weight can decrease but its voting power increase, and vice-versa. More recently, Leech (1998) explor ed a way to estimate voting power values for large committee systems, such as the Board of Governors of the IMF.

(4.) For a highly personal, yet informative, account of the admission of non-regional members, see Fordwor (1981). Shaw (1991) in reference to the debate regarding the admission of non-regional members describes how "[t]hrough bargaining, the relatively small countries of Algeria and Libya aligned with Nigeria in keeping the non-regional members at bay, against the wishes of 41 other African states, until domestic forces in Nigeria favouring the opening of capital upset the balance" (p. 543, passim).

(5.) On the question of how the admission of non-regional members affected policy outputs, see Mingst (1990:60-66).

(6.) Annex B, "The Election of Directors," in the first finalized edition of the Articles of Agreement.

(7.) A special majority of two-thirds of the members holding at least 75 percent of the voting weight is required to change the ADB's Articles of Agreement. Further, such a coalition must include of two-thirds of the regional members holding at least 75 percent of the regional members' voting weight.

(8.) Compare the ADB's 625 basic votes to the IDB's 135 and the World Bank's 250. The AsDB uses a different rational for assigning basic votes. It divides 20 percent of all votes evenly across members.

(9.) Other early work on voting power include Banzhaf (1965; 1969); Brams and Affuso (1976); Deegan and Packel (1978); Johnston (1978); Dubey and Shapley (1979); and Fischer and Schotter (1978).

(10.) While voting power indices have been applied in interesting ways there is no single index that can claim to be the "best" measure of voting power. The lack of a universally acceptable measure of power has engendered some criticisms of voting-power indices in part because results across indices are often not consistent (e.g., Felsenthal et al. 1998). For more perspectives on the utility of power indices see Felsenthal and Machover (1995); Garrett and McLean (1996); Garrett et al. (1995); Holler (1997); Johnston (1996; 1995a; 1995b); Morriss (1996; 1987); Nurmi (1997); Tsebelis and Garrett (1996); and Turnovec (1997). Riker (1992) provides a useful account of the use of power indices in political science. Van Deemen (1989) offers an interesting examination of the role of critical actors in coalition formation. The most comprehensive analysis of power indices is found in Felsenthal and Machover (1998). Garrett and Tsebelis (1999) and Lane and Berg (1999) have recently debated pros and cons of voting power indices in the context of the European Union.

(11.) Voting games are generally assumed to be proper, i.e., q [greater than] 1/2 [[[sigma].sup.n].sub.i=1] [w.sub.i] a condition that there cannot be two disjoint winning coalitions.

(12.) There are, of course, other ways to divide-up points for being a critical member. For example, a power index could assign each critical player the value of 1/n where n is the total number of players in the coalition.

(13.) Owen (1977; 1981) does not present results for the Johnston index, but the multilinear method should apply just as well to the Johnston index as it does to the Banzhaf and Shapley-Shubik indices.

(14.) Guinea Bissau, Equatorial Guinea, and Mauritius did not participate in the 1975 election of members of the Board of Directors and thus their votes are excluded from the analysis.

(15.) Angola, Equatorial Guinea and Zimbabwe did not take part in the 1980 election of members to the Board of Directors and are thus excluded from the analysis.

(16.) South Africa and Eritrea were not represented in the Board of Directors and are therefore excluded from the analysis.

(17.) There are 6 non-regional representatives to the Board of Directors but the United States appoints its representative and therefore there are only 5 non-regional voting groups.

(18.) Japan's behavior, in contrast to its dictator status, is described by Shaw (1991) as "a generous, quiet uncle... Japanese support... has been characterized by substantial outlays of cash, unfettered by conditions and demands of any sort" (p. 545).

(19.) A large member in a voting group would want to have its number of votes equal to all the other members of the group plus one. This would maximize the number of votes the large member could capture as a dictator.

(20.) This quote is the title of Banzhaf (1969).

REFERENCES

AFRICAN DEVELOPMENT BANK

Articles of Agreement. Abidjan, Ivory Coast.

Annual Report, various years. Abidjan, Ivory Coast.

ASH, M.A.

1951 "An Analysis of Power, with Special Reference to International Politics." World Politics 3(2):218-37.

BALDWIN, D.A.

1989 Paradoxes of Power. New York: Basil Blackwell.

BANZHAF, J.F. III

1969 "Weighted Voting Doesn't Work: A Mathematical Analysis." Rutgers Law Review 19:317-343.

1965 "One Man, 3,312 Votes: A Mathematical Analysis of the Electoral College." Villanova Law Review 13:304-32.

BRAMS, S.J.

1990 Negotiation Games: Applying Game Theory to Bargaining and Arbitration. New York: Routledge.

1985 Rational Politics: Decisions, Games, and Strategy. Washington, D.C.: Congressional Quarterly Press.

1978 The Presidential Election Game. New Haven: Yale University Press.

BRAMS, S.J. and P.J. AFFUSO

1976 "Power and Size: A New Paradox." Theory and Decision 7:29-56.

BRAMS, S.J., P.J. AFFUSO, and D.M. KILGOUR

1989 "Presidential Power: A Game-Theoretic Analysis." Pp. 55-74 in The Presidency in American Politics, edited by P. Bruce and G. King. New York: NYU Press.

DEEGAN, J. and E. PACKEL

1978 "A New Index for Simple n-Person Games." International Journal of Game Theory 7:113-123.

DIXON, W.J.

1983 "The Evaluation of Weighted Voting Schemes for the United Nations General Assembly." International Studies Quarterly 27:295-314.

DREYER, J. and A. SCHOTTER

1980 "Power Relationships in the International Monetary Fund: The Consequences of Quota Changes." Review of Economics and Statistics 62(1):97-106.

DUBEY, P. and L. SHAPLEY

1979 "Mathematical Properties of the Banzhaf Power Index." Mathematical Operational Resources 4:99-131.

EISNER, M.P.

1992 "Estimating Voting Power in Presidential Elections." Mathematics and Computer Education 21(1):50-59.

FELSENTHAL, D.S. and M. MACHOVER

1998 The Measurement of Voting Power: Theory and Practice, Problems and Paradoxes. Cheltenham, UK: Edward Elgar.

1995 "Postulates and Paradoxes of Relative Voting Power: A Critical Re-Appraisal." Theory and Decision 38(2):195-229.

FELSENTHAL, D.S., M. MACHOVER, and W. ZWICKER

1998 "The Bicameral Postulates and Indices of A Priori Voting Power." Theory and Decision 44:83-116.

FISCHER, D. and A. SCHOTTER

1978 "The Inevitability of the Paradox of Redistribution in the Allocation of Voting Weights." Public Choice 33(2):49-67.

FORDWOR, K.D.

1981 The African Development Bank: Problems of International Cooperation. New York: Pergamon.

FREIXAS, J. and G. GAMBARELLI

1997 "Common Internal Properties Among Power Indices." Control and Cybernetics 26(4):591-603.

GAMBARELLI, G.

1994 "Power Indices for Political and Financial Decisionmaking: A Review." Annuals of Operations Research 51:165-73.

GARRETT, G. and I. MCLEAN

1996 "On Power Indices and Reading Papers." British Journal of Political Science 26:600.

GARRETT, G., I. MCLEAN, and M. MACHOVER

1995 "Power, Power Indices and Blocking Power: A Comment on Johnston." British Journal of Political Science 25(4):563-568.

GARRETT, G. and G. TSEBELIS

1999 "Why Resist the Temptation to Apply Power Indices to the European Union?" Journal of Theoretical Politics 11(3):291-309.

1996 "An Institutional Critique of Intergovernmentalism." International Organization 50(2):269-99.

GIANARIS, William N.

1990 "Weighted Voting in the International Monetary Fund and the World Bank." Fordham International Law Journal 14(4):910-45.

HOLLER, M.J.

1997 "Power, Monotonicity, and Expectations." Control and Cybernetics 26(4):605-07.

HOSLI, M.O.

1997 "Voting Strength in the European Parliament: The Influence of National and of Partisan Actors." European Journal of Political Research 31:351-66.

JOHNSTON, R.J.

1996 "On Keeping Touch with Reality and Failing to be Befuddled by Mathematics." British Journal of Political Science 26:598-99.

1995a "The Conflict over Qualified Majority Voting in the European Union Council of Ministers: An Analysis of the UK Negotiation Stance Using Power Indices." British Journal of Political Science 25(2):245-254.

1995b "Can Power Be Reduced to a Quantitative Index -- and If So, Which One? A Response to Garrett, McLean and Machover." British Journal of Political Science 24(5):568-72.

1978 "On the Measurement of Power: Some Reactions to Laver." Environment and Planning 10A:907-914.

KILGOUR, D.M.

1985 "Reply: Distributing the Power to Amend Canada's Constitution." Canadian Journal of Political Science 18(2):389-96.

1983 "A Formal Analysis of the Amending Formula of Canada's Constitution Act, 1982." Canadian Journal of Political Science 16(4):771-77.

LAMBERT, J.P.

1988 "Voting Games, Power Indices, and Presidential Elections." UMAP Journal 9(3):213-67.

LANE, J. and S. BERG

1999 "Relevance of Voting Power." Journal of Theoretical Politics 11(3):309-20.

LEECH, D.

1998 "Power Relations in the International Monetary Fund: A Study of the Political Economy of A Priori Voting Power using the Theory of Simple Games." Warwick Economic Research Papers No. 494, Department of Economics, University of Warwick.

1992 "Empirical Analysis of the Distribution of A Priori Voting Power: Some Results for the British Labour Party Conference and Electoral College." European Journal of Political Research 21:245-65.

1990 "Power Indices and Probabilistic Voting Assumptions." Public Choice 66:293-99.

LEVESQUE, T.J. and J.W. MOORE

1984 "Citizen and Provincial Power Under Alternative Amending Formulae: An Extension of Kilgour's Analysis." Canadian Journal of Political Science 17(1):157-66.

MERCIK, J.W.

1997 "Power and Expectations." Control and Cybernetics 26(4):617-21.

MINGST, K.A.

1990 Politics and the African Development Bank. Lexington: University of Kentucky Press.

MORRISS, P.

1996 "Qualified Majority Voting and Power Indices: A Further Response to Johnston." British Journal of Political Science 26:595-97.

1987 Power: A Philosophical Analysis. Manchester: Manchester University Press.

NEWCOMBE, H., J. WERT, and A. NEWCOMBE

1971 "Comparison of Weighted Voting Formulas for the United Nations." World Politics 23(3):452-492.

NURMI, H.

1997 "On Power Indices and Minimal Winning Coalitions." Control and Cybernetics 26(4):609-611.

O'NEILL, B.

1996 "Power and Satisfaction in the United Nations Security Council." Journal of Conflict Resolution 40(2):219-37.

OWEN, G.

1981 "Modification of the Banzhaf-Coleman Index for Games with A Priori Unions." Pp. 232-238 in Power, Voting, and Voting Power, edited by M.J. Holler. Wurzburg: Physica

1977 "Values of Games with A Priori Unions." Pp. 76-88 in Mathematical Economics and Game Theory, edited by R. Henn and O. Moeschlin. Berlin: Springer.

RIKER, W.H.

1992 "The Entry of Game Theory into Political Science." Pp. 207-223 in Toward a History of Came Theory, edited by E. Roy Weintraub. Durham and London: Duke University Press.

SCHOTTER, A.

1981 "The Paradox of Redistribution: Some Theoretical and Empirical Results." Pp. 324-328 in Powr, Voting, and Voting Power, edited by M.J. Holler. Wurzburg: Physica-Verlag.

SHAPLEY, L.S. and M. SHUBIK

1954 "A Method for Evaluation the Distribution of Power in a Committee System." American Political Science Review 48:787-792.

SHAW, C.L.

1991 "Par Inter Paribus: The Nature of Power in Cooperation. Lessons (For the United States) from the African Development Bank." African Affairs 90:537-58.

STRAND, J.R.

1999 "State Power in a Multilateral Context: Voting Strength in the Asian Development Bank." International Interactions 25(3):53-74.

STRANGE, S.

1996 The Retreat of the State: The Diffusion of Power in the World Economy. New York: Cambridge University Press.

TSEBELIS, G. and G. GARRETT

1996 "Agenda Setting Power, Power Indices, and Decision Making in the European Union." International Review of Law and Economics 16(3):345-61.

TURNOVEC, F.

1997 "Power, Power Indices, and Intuition." Control and cybernetics 26(4):613-15.

VAN DEEMEN, A.M.A.

1989 "Dominant Players and Minimum Size Coalitions." European Journal of Political Research 17:313-32.
Table 1

Voting Strength in the ADB, 1975


Country        VW      Bm      Bs      Bt      Jm      Js      Jt

Nigeria [*]  0.0962  1.0000  0.1327  0.1327  1.0000  0.1621  0.1621
Algeria [*]  0.0697  1.0000  0.1327  0.1327  1.0000  0.1621  0.1621
Libya        0.0688  0.3333  0.1327  0.0442  0.3333  0.1621  0.0540
Egypt        0.0620  0.3333  0.1327  0.0442  0.3333  0.1621  0.0540
Upper Volta  0.0129  0.3333  0.1327  0.0442  0.3333  0.1621  0.0540
Ethiopia     0.0283  0.3333  0.1003  0.0334  0.5333  0.0856  0.0457
Senegal      0.0201  0.3333  0.1003  0.0334  0.5333  0.0856  0.0457
Morocco      0.0405  0.3333  0.1003  0.0334  0.3333  0.0856  0.0285
Zambia       0.0357  0.3333  0.1003  0.0334  0.3333  0.0856  0.0285
Ghana        0.0343  0.3333  0.1003  0.0334  0.3333  0.0856  0.0285
Sudan        0.0280  0.3333  0.1003  0.0334  0.3333  0.0856  0.0285
Tunisia      0.0242  0.3333  0.1003  0.0334  0.3333  0.0856  0.0285
Malawi       0.0148  0.3333  0.1003  0.0334  0.3333  0.0856  0.0285
Tanzania     0.0269  0.2857  0.1003  0.0287  0.3333  0.0856  0.0285
Ivory Coast  0.0249  0.2857  0.1003  0.0287  0.3333  0.0856  0.0285
Kenya        0.0249  0.2857  0.1003  0.0287  0.3333  0.0856  0.0285
Cameroon     0.0134  0.2857  0.1003  0.0287  0.3333  0.0856  0.0285



(*)Denotes dictator.
Table 2

Voting Strength in the ADB, 1980


Country        VW      Bm      Bs      Bt      Jm      Js      Jt

Nigeria [*]  0.1169  1.0000  0.1182  0.1182  1.0000  0.1281  0.1281
Libya [*]    0.0709  1.0000  0.1182  0.1182  1.0000  0.1281  0.1281
Egypt        0.0605  0.6364  0.1182  0.0752  0.8000  0.1281  0.1025
Algeria [*]  0.0828  1.0000  0.1182  0.1182  1.0000  0.0992  0.0992
Morocco      0.0412  0.6364  0.1054  0.0671  0.8000  0.0992  0.0794
Ivory Coast  0.0293  0.3448  0.1054  0.0364  0.4872  0.0992  0.0484
Zambia       0.0375  0.3333  0.1182  0.0394  0.3333  0.1094  0.0365
Ghana        0.0350  0.3333  0.1182  0.0394  0.3333  0.1094  0.0365
Mozambique   0.0175  0.3333  0.1182  0.0394  0.3333  0.1094  0.0365
Senegal      0.0161  0.2422  0.1054  0.0255  0.3085  0.0992  0.0306
Gabon        0.0155  0.2422  0.1054  0.0255  0.3085  0.0992  0.0306
Kenya        0.0251  0.2258  0.1182  0.0267  0.2557  0.1094  0.0280
Ethiopia     0.0247  0.2258  0.1182  0.0267  0.2557  0.1094  0.0280
Tanzania     0.0231  0.1935  0.1182  0.0229  0.1970  0.1094  0.0215
Cameroon     0.0231  0.2069  0.1054  0.0218  0.2058  0.0992  0.0204
Burundi      0.0089  0.1641  0.1054  0.0173  0.1649  0.0992  0.0164
Uganda       0.0135  0.1290  0.1182  0.0153  0.1098  0.1094  0.0120
Mauritius    0.0128  0.1290  0.1182  0.0153  0.1098  0.1094  0.0120



(*)Denotes dictator.
Table 3

Voting Strength in the ADB, 1985


                 VW      Bm      Bs      Bt      Jm      Js      Jt

Japan          0.0471  0.8611  0.0991  0.0853  0.8871  0.2375  0.2107
Nigeria [*]    0.0877  1.0000  0.0886  0.0886  1.0000  0.1644  0.1644
Germany [*]    0.0357  1.0000  0.0611  0.0611  1.0000  0.0496  0.0496
France [*]     0.0326  1.0000  0.0600  0.0600  1.0000  0.0470  0.0470
Egypt [*]      0.0581  1.0000  0.0596  0.0596  1.0000  0.0466  0.0466
US             0.0569  1.0000  0.0559  0.0559  1.0000  0.0412  0.0412
Cote-d'Ivoire  0.0254  0.5417  0.0620  0.0336  0.7184  0.0524  0.0376
Canada [*]     0.0326  1.0000  0.0521  0.0521  1.0000  0.0367  0.0367
Morocco [*]    0.0318  1.0000  0.0516  0.0516  1.0000  0.0362  0.0362
Zaire [*]      0.0303  1.0000  0.0490  0.0490  1.0000  0.0338  0.0338
Zambia [*]     0.0254  1.0000  0.0445  0.0445  1.0000  0.0300  0.0300
Algeria [*]    0.0270  1.0000  0.0393  0.0393  1.0000  0.0257  0.0257
Ghana [*]      0.0254  1.0000  0.0392  0.0392  1.0000  0.0256  0.0256
Zimbabwe       0.0269  0.5000  0.0539  0.0270  0.6429  0.0387  0.0249
Libya [*]      0.0363  1.0000  0.0364  0.0364  1.0000  0.0236  0.0236
Ethiopia       0.0189  0.3462  0.0610  0.0211  0.4100  0.0493  0.0202
Tanzania       0.0135  0.2692  0.0610  0.0164  0.2600  0.0493  0.0128
Kenya          0.0162  0.2692  0.0610  0.0164  0.2600  0.0493  0.0128
Switzerland    0.0134  0.2333  0.0566  0.0132  0.2652  0.0422  0.0112
Sweden         0.0141  0.2333  0.0566  0.0132  0.2652  0.0422  0.0112



(*)Denotes dictator.
Table 4

Voting Strength in the ADB, 1990


Country        VW      Bm      Bs      Bt      Jm      Js      Jt

Nigeria [*]    0.1060  1.0000  0.1120  0.1120  1.0000  0.3487  0.3487
Egypt [*]      0.0615  1.0000  0.0619  0.0619  1.0000  0.0569  0.0569
Germany [*]    0.0362  1.0000  0.0597  0.0597  1.0000  0.0474  0.0474
France [*]     0.0330  1.0000  0.0597  0.0597  1.0000  0.0472  0.0472
Japan [*]      0.0485  1.0000  0.0587  0.0587  1.0000  0.0452  0.0452
US             0.0593  1.0000  0.0584  0.0584  1.0000  0.0446  0.0446
Morocco [*]    0.0358  1.0000  0.0557  0.0557  1.0000  0.0403  0.0403
Cote d'Ivoire  0.0280  0.8378  0.0557  0.0467  0.9365  0.0404  0.0379
Libya [*]      0.0441  1.0000  0.0505  0.0505  1.0000  0.0365  0.0365
Canada [*]     0.0333  1.0000  0.0513  0.0513  1.0000  0.0354  0.0354
Zaire [*]      0.0428  0.0000  0.0505  0.0000  1.0000  0.0347  0.0347
Zimbabwe       0.0281  0.5000  0.0595  0.0297  0.6429  0.0467  0.0300
Algeria [*]    0.0363  1.0000  0.0449  0.0449  1.0000  0.0298  0.0298
Zambia [*]     0.0261  1.0000  0.0443  0.0443  1.0000  0.0294  0.0294
Ghana [*]      0.0238  1.0000  0.0505  0.0505  1.0000  0.0231  0.0231
Ethiopia       0.0194  0.7500  0.0397  0.0298  0.8871  0.0258  0.0229
Sweden         0.0141  0.2333  0.0548  0.0128  0.2652  0.0393  0.0104
Switzerland    0.0133  0.2333  0.0548  0.0128  0.2652  0.0393  0.0104



(*)Denotes dictator.
Table 5

Voting Strength in the ADB, 1995


Country              VW      Bm      Bs      Bt      Jm      Js

Nigeria [*]        0.1014  1.0000  0.1091  0.1091  1.0000  0.2902
Zimbabwe           0.0268  0.3462  0.0802  0.0278  0.4100  0.1286
Japan [*]          0.0470  1.0000  0.0611  0.0611  1.0000  0.0496
Egypt [*]          0.0586  1.0000  0.0608  0.0608  1.0000  0.0483
Germany [*]        0.0354  1.0000  0.0599  0.0599  1.0000  0.0464
France [*]         0.0323  1.0000  0.0598  0.0598  1.0000  0.0460
Morocco [*]        0.0380  1.0000  0.0578  0.0578  1.0000  0.0426
US                 0.0570  1.0000  0.0575  0.0575  1.0000  0.0422
Canada [*]         0.0323  1.0000  0.0564  0.0564  1.0000  0.0407
Algeria [*]        0.0408  1.0000  0.0504  0.0504  1.0000  0.0345
Botswana           0.0241  0.2692  0.0802  0.0216  0.2600  0.1286
Angola             0.0132  0.2692  0.0802  0.0216  0.2600  0.1286
Cote d'Ivoire [*]  0.0410  1.0000  0.0479  0.0479  1.0000  0.0315
Libya [*]          0.0386  1.0000  0.0444  0.0444  1.0000  0.0292
Ghana [*]          0.0229  1.0000  0.0418  0.0418  1.0000  0.0271
Zaire [*]          0.0210  1.0000  0.0385  0.0385  1.0000  0.0246
Ethiopia           0.0185  0.3462  0.0547  0.0189  0.4100  0.0388
Zambia             0.0135  0.5000  0.0315  0.0158  0.6429  0.0199
Kenya              0.0169  0.2692  0.0547  0.0147  0.2600  0.0388
Tanzania           0.0098  0.2692  0.0547  0.0147  0.2600  0.0388
Sweden             0.0135  0.2333  0.0539  0.0126  0.2652  0.0379
Switzerland        0.0128  0.2333  0.0539  0.0126  0.2652  0.0379




Country              Jt

Nigeria [*]        0.2902
Zimbabwe           0.0527
Japan [*]          0.0496
Egypt [*]          0.0483
Germany [*]        0.0464
France [*]         0.0460
Morocco [*]        0.0426
US                 0.0422
Canada [*]         0.0407
Algeria [*]        0.0345
Botswana           0.0334
Angola             0.0334
Cote d'Ivoire [*]  0.0315
Libya [*]          0.0292
Ghana [*]          0.0271
Zaire [*]          0.0246
Ethiopia           0.0159
Zambia             0.0128
Kenya              0.0101
Tanzania           0.0101
Sweden             0.0100
Switzerland        0.0100



(*)Denotes dictator.
Appendix A

Voting Groups and Their Voting Weights in the African Development
Bank, 1995.

Below is a list of members of the African Development Bank in 1995


                                                  Total Votes

Member
United States (appointed)                            90077

Voting Group
Cote d'Ivoire, Equatorial Guinea, Guinea             74705
Ghana, Gambia, Liberia, Sierra Leone, Sudan          66684
Algeria, Madagascar, Guinea Bissau                   79075
Morocco, Tunisia, Togo                               90499
Zambia, Lesotho, Malawi, Mauritius, Swaziland        50626
Gabon, Benin, Burkina Faso, Cape Verde, Comoros,     55142
 Mali, Niger, Senegal, Chad
Egypt, Djibouti                                      94787
Zaire, Cameroon, Congo, CAR                          60886
Uganda, Tanzania, Ethiopia, Kenya, Rwanda,           86023
 Seychelles
Libya, Mauritania, Somalia                           70505
Mozambique, Botswana, Angola, Burundi, Namibia,     125953
Zimbabwe
Nigeria, Sao Tome & Principe                        162604
Sweden, Finland, Denmark, India, Norway,             84798
 Switzerland
Italy, France, Belgium                               93387
Japan, Saudi Arabia, Argentina, Austria, Brazil      95405
Germany, Netherlands, Portugal, UK                   93648
Canada, Spain, China (PRC), Korea, Kuwait            88381




                                                  Voting Weight

Member
United States (appointed)                             5.69

Voting Group
Cote d'Ivoire, Equatorial Guinea, Guinea              4.73
Ghana, Gambia, Liberia, Sierra Leone, Sudan           4.22
Algeria, Madagascar, Guinea Bissau                    4.93
Morocco, Tunisia, Togo                                5.72
Zambia, Lesotho, Malawi, Mauritius, Swaziland         3.2
Gabon, Benin, Burkina Faso, Cape Verde, Comoros,      3.49
 Mali, Niger, Senegal, Chad
Egypt, Djibouti                                       5.99
Zaire, Cameroon, Congo, CAR                           3.85
Uganda, Tanzania, Ethiopia, Kenya, Rwanda,            5.44
 Seychelles
Libya, Mauritania, Somalia                            4.46
Mozambique, Botswana, Angola, Burundi, Namibia,       7.97
Zimbabwe
Nigeria, Sao Tome & Principe                         10.28
Sweden, Finland, Denmark, India, Norway,              5.36
 Switzerland
Italy, France, Belgium                                5.91
Japan, Saudi Arabia, Argentina, Austria, Brazil       6.03
Germany, Netherlands, Portugal, UK                    5.92
Canada, Spain, China (PRC), Korea, Kuwait             5.59



Source: African Development Bank Annual Report, 1995.

Note that South Africa and Eritrea did not participate in the
election of Executive Directors so their votes, 15635 and 1039
respectively, are not listed.
COPYRIGHT 2001 Sage Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:STRAND, JONATHAN R.
Publication:Journal of Asian and African Studies
Date:May 1, 2001
Words:9630
Previous Article:The Transformation of Sri Lankan Society, 1948-1999: The Fragmentation of Centralism [1].
Next Article:Post-Cold War Realism, Liberal Internationalism, and the Third World.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters