THERE is a widespread agreement that the International Monetary (IMF) has played a central role in fostering a dynamic and open international economy in the postwar period. As the world economy evolved, the consensus has also emerged, however, that substantial changes were required in the Fund's rules so as to make them better reflect the changing power relationships among its members. In an attempt to make these changes, the Proposed Second Amendment to the Articles of Agreement of the International Monetary Fund (IMF, 1976) was offered to the IMF's members for ratification along with a change in the distribution of the members' quotas. The new rules became effective as of April 1, 1978, when the amendment was ratified by the required 60% of members with 80% of the total vote. It is the purpose of this paper to demonstrate, using some simple game theoretical tools, that in many cases the framers of these amendments and quota changes have achieved a result that may be exactly the opposite of their intentions. More specifically, we will show that 1) Under the proposed changes 38 countries in the IMF have their percentage of the total vote decreased and yet their voting power within the organization increased when power is measured by the Banzhaf power index. 2) With the new vote weight distribution and voting rules, four major countries-Belgium, Holland, West Germany and Japan have their percentage of the total vote increased and yet have their percentage of the total power decreased. Paradoxical results of this variety were discussed theoretically by Fischer and Schotter (1978) and labeled the Paradox of Redistribution.' 1 3) Under the previous distribution of votes and voting rules, smaller countries had voting powers that were out of proportion to their voting weights, and the newly introduced changes would generally aggravate this disproportion. 4) The power of the United States within the increases substantially on issues where most countries vote through their Executive Directors (as groups), as opposed to issues where they vote through their Governors (individually). 5) Under both the previous and current voting system, significant diminishing returns to voting weights exist and the tendency is more pronounced under the new system. In most voting situations power is a concave function of voting weight with large linear segments. 6) Although these results are in many instances not quantitatively substantial, qualitatively they indicate a noticeable discrepancy between what one would think the consequences of the voting changes would be and what they actually are. In this paper we will proceed as follows: Section I will present some background material about the voting procedures at the IMF. Section II will discuss the index that we are using to measure power. Section III will present the results of our calculations and discuss their significance. Finally, section IV will offer some conclusions and consider the relevance of our findings.
Read full abstract