In response to the current worldwide financial and economic crisis, the so-called "G20" group has emerged as the premier forum of economic policymakers. Since its first meeting in November 2008, the G20 has replaced the G8 group of large industrialised economies, which had originally been established in the 1970s as the G5. The expansion from 8 to 20 members reflects the shift in global economic weight from the established industrialised countries to emerging economies.1 Thus, it can be assumed that the G20 is closer to a globally shared opinion on economic affairs than the G8 had been. This should give the G20 more legitimacy and leverage in implementing its agenda, an agenda which includes a reform of IMF governance. This rise in relevance formed the impetus for our analysis of the proposal from the world's leading policy forum (G20) for the world's leading international financial institution (IMF).
The reform of IMF governance is not just one issue among a set of reforms but a crucial focal point. The IMF is centre stage whenever proposals for stabilising the international financial architecture are made because there is no other comparable institution. If, however, the IMF is so important for any reform agenda, then its governance is at the heart of such an agenda. In a sense, IMF governance is of paradigmatic importance: whatever a proposal wants to achieve through the reform of international financial institutions will be reflected in its stance on IMF governance. Accordingly, major interests and contradictory ideas are involved in the debate about IMF governance. This makes the G20's decision on this issue even more important. We proceed with our analysis in three steps. First, we report what has been decided. Second, as a prerequisite to understanding much of the heated debate, we introduce the IMF's quota system, which determines the voting power of nations within the IMF. Third, we provide an assessment of the G20 reform proposal.
The G20's Guideline
Before going into detail, it is worth reflecting on the spirit that pervades the communiqués published after each of the three G20 summits that have taken place so far.2 It is quite obvious that the "new" members of this international forum are pushing for broader participation of emerging and developing countries. These members have been successful, as the G20 acknowledges a lack of legitimacy and ownership of international financial institutions, including the IMF. For instance, the London Communiqué states: "We will reform their [the international financial institutions'] ... governance to reflect changes in the world economy and the new challenges of globalisation, and that emerging and developing economies, including the poorest, must have greater voice and representation."3
"Voice" and "representation", "legitimacy" and "ownership" are keywords, indicating that most countries of the world do not see the IMF as "their" institution. Instead, the IMF is correctly regarded as the instrument of a small group of industrialised countries, in particular the United States and the European Union.4 This perception damages the IMF's operations: if debtor countries have the impression that the policies and conditions they are obliged to fulfil have been constructed to pursue goals other than their economic wellbeing, the legitimacy and credibility of IMF programmes will be undermined. Consequently, countries will be more reluctant to accept these conditions and may even hesitate to ask for IMF aid.5
The G20 Proposal of Four Measures
In light of the IMF's acknowledged shortcomings, the G20 has agreed on four measures towards improving IMF governance:
- shifting at least 5% of votes from over-represented to under-represented (emerging market and developing) countries;6
- accelerating the next general quota review;7
- delinking the election of the IMF's managing director from regional origin;8
- putting forward the Singapore quota and voice reform.9
We shall now discuss these measures.
(1) Vote Shifting
The G20 agreed to a 5% shifting of votes towards under-represented countries, i.e. towards emerging and developing countries. The 5% shift aims to better reflect "members' relative weights in the world economy"10 with regard to vote distribution. As votes are mainly determined by quotas, the G20 refers to "using the current quota formula as the basis to work from."11 This seems reasonable because the IMF's system of quota formulas, i.e. the Calculated Quota System (CQS), was designed "to guide the assessment of a member's relative position."12 Consequently, the deviation of a country's actual quota share from its CQS will be used as the benchmark for under- and over-representation for an evaluation of a five per cent shift of quotas.
(2) Antedated General Quota Review
The Fund's Articles oblige the IMF to perform general quota reviews every five years.13 The last review was finished in early 2008. Thus, the next review needs to be completed by 2013. In light of the need for a quota adjustment, G20 leaders called for the acceleration of the forthcoming review with the goal of finishing it in January 2011, i.e. at least two years ahead of schedule. However, in light of the "difficult and time-consuming"14 negotiations, the narrow time-frame of less than two years from the time the decision was made at the London Summit in March 2009 until the proposed completion of negotiations in January 2011 is very ambitious.
(3) Election of the Managing Director
With respect to the US-EU consensus on choosing the IMF's managing director,15 the London Communiqué and the Pittsburgh Leaders' Statement stress that future criteria for electing the heads of all international institutions need to be based upon qualifications rather than regional origin.16 This decision marks a turning-point in the Fund's history and follows the unanimous recommendations of the G20 Working Group 3,17 the Stiglitz Commission,18 the Manuel Group,19 the IEO20 and many critics.21 The decision to expand the proposal of an open and merit-based election not only to the IMF but to all international institutions is wise. EU countries would hardly support a reformed election process for the Fund if the US continued to claim the World Bank presidency for an American. However, the G20 communiqués do not provide details of such a process or the criteria included, nor do they determine a deadline for implementing these processes. The terms of both institutions' heads (Mr. Zoellick as president of the World Bank and Mr. Strauss-Kahn as the Fund's managing director) continue through 2012. However, at the time of his election, it was already presumed that Mr. Strauss-Kahn would be the last European to hold the office of managing director.22
(4) Implementing the Singapore Reform
The last measure that was decided by G20 leaders with respect to governance reform is their support for the Singapore quota and voice reform. While the London Communiqué stated the commitment of G20 countries to implement the Singapore reforms, the call for implementation of the reform package was much more pressing at the most recent Pittsburgh Summit: "We must urgently implement the package of IMF quota and voice reforms agreed in April 2008."23
In April 2008 the IMF's Board of Governors approved a quota and voice reform package that was first discussed by members at the Fund's 2006 annual meeting in Singapore. This package includes five elements (see Table 1). In the meantime, one of these elements has already been implemented. The first round ad hoc quota increase was realised in 2006 and aimed at levelling the imbalances of the four countries that were the "most underrepresented members"24, i.e. China, Korea, Mexico and Turkey. It distributed SDR 3.81 billion of quotas, i.e. 1.8% of IMF's total quotas.25
Table 1 (back to the text)
Elements of the Singapore Reform Package
|first round ad hoc increase||initial (first round) ad hoc increase in quotas for the most under-represented members: China, Korea, Mexico, and Turkey|
|new quota formula||replacing the current five-formula system by one new quota formula to achieve a more simple and transparent calculation that better reflects countries' relative economic weight|
|second round ad hoc increase||second round of ad hoc quota increases based on the new formula|
|additional alternate executive directors||additional alternate executive directors for the two African chairs represented on the IMF's Executive Board|
|tripling of basic votes||tripling of basic votes to counteract the erosion of basic votes' importance|
A second element that has already been partially implemented is the appointment of additional staff (but not alternate executive directors yet) for each of the two constituencies representing African members.26 According to the IMF, the heavy workload of these constituencies, caused by the large number of represented countries and the importance of IMF funding for these countries, justifies this decision. Indeed, by number of countries, they are the largest constituencies, representing 20 and 23 members of the Executive Board respectively (although this adds up to only 3.01% and 1.35% of total votes). Hence, the decision seems reasonable but is only a minor aspect of the reforms.
Beyond these two elements, the "second round" ad hoc quota increases based on the new formula are picked up by the 5% quota reallocation endorsed by the G20. Thus, there are still two open elements of the Singapore reform package which refer to the redistribution of voting power: the creation of a new quota formula and the tripling of basic votes. Discussion of the quota formula will require an understanding of the IMF quota system.
The IMF Quota System
The quota which each IMF member country receives has four functions: first, it determines access to IMF funding; second, it determines obligations to the IMF; third, it determines the number of votes in the Board of Governors (in combination with the basic votes); and fourth, it determines the share of Special Drawing Right (SDR) allocations. The diverse factors that make quotas important are reflected in the methodology for calculating the quotas.
The concrete formula for quota calculation presently consists of five sub-formulas which contain similar elements that are assigned different weights.27 As a result, the formula looks quite complicated, and it is not at all obvious why it has been constructed in such a manner. The only explanation for its present form lies in the process of constructing the formula. IMF member countries had to fight for their interests by pushing for certain formula elements that were favourable to their country. Despite a process dominated by political compromises, the logic behind the inclusion of specific elements in the formula is recognisable. In detail, there are five elements that are taken into consideration, as can be seen from Table 2.
Table 2 (back to the text)
Elements of Current Quota Formulas (the CQS)
|Y||GDP at current market prices|
|R||twelve-month average of gold, foreign exchange reserves, SDR holdings and reserve positions in the IMF|
|P||annual average of current payments (goods, services, income and private transfers) for a recent five-year period|
|C||annual average of current receipts (goods, services, income and private transfers) for a recent five-year period|
|V||variability of current receipts, defined as one standard deviation from the centred five-year moving average, for a recent 13-year period|
The element "GDP" reflects the size of an economy, "reserves" reflect the country's ability to support IMF financing, "current payments" and "current receipts" are considered to account for openness and thus for relevance to IMF operations, and the "variability of receipts" is seen as a measure of vulnerability and thus of a potential need for IMF support. Overall, the consideration of these elements does not seem to be inappropriate, although their weighting is open to discussion.
Figure 1 (back to the text)
15 Most Under-represented and 15 Most Over-represented Countries Measured by Percentage Point Difference of Actual Quota Share from CQS
Calculatory Quotas as Benchmark
If we now take the present CQS as a benchmark for the shift in power as envisioned by the G20, we receive a surprising result. Figure 1 shows the 15 most over-represented and 15 most under-represented countries by plotting the difference between the CQS and the actual quota share based on 2006 data (which still reflects the current status). Singapore, for example, currently holds approximately 0.4% of quotas, while the calculated quotas would entitle it to approximately 1.9%; China, which currently holds approximately 3.7% of quotas, would hold 5.2% under the CQS. Astonishingly, more than half of the 15 most under-represented countries are EU members (8 out of 15). At the top of Figure 1, the most over-represented countries are listed. For instance, Saudi Arabia currently holds more than three times the quotas of its calculated value (3.2% instead of 1%); Russia holds approx. 2.7% of quotas, which is also considerably more than its calculated share of approximately 1.5%.
Overall, if this definition were to be used for levelling quota shares, it would be mainly EU countries and not developing countries (apart from China or Korea, for example) which would benefit from a 5% shift. Following the IMF's classification of developing and emerging economies,28 70% of all developed countries (22 out of 31) are under-represented while more than three-quarters (117 out of 152) of the emerging and developing countries are over-represented. Hence, if the G20 leaders' aim is to give greater voice to developing countries, other indicators of under-representation need to be used. A starting point may be the new formula recently suggested by the IMF.29
New IMF Quota Formula
Interestingly, an overhaul of the quota calculation process was agreed upon, although it has not yet been implemented. This was to be achieved via the introduction of a new quota formula that supersedes the current system of five formulas. The aim of this new formula is a simpler and more transparent calculation that better reflects countries' relative economic weights.30 The new formula relies on the same elements as the present formula but reduces the number of sub-formulas to one.31 Moreover, it integrates current payments and receipts into a single measure of openness and modifies the conversion weights for GDP calculation from 100% market rates to a combination of 60% market and 40% PPP exchange rates (see Table 3). This latter modification increases the weight of developing countries. Finally, the compression factor was mainly supported by small EU countries in order to offset the high correlation of size-related variables.32
Table 3 (back to the text)
New Quota Formula
|CQS = (0.5Y+0.3O+0.15V+0.05R)k|
|with||CQS||=||Calculated Quota Share|
|Y||=||GDP converted at market rates (weight 60%) and PPP exchange rates (weight 40%) averaged over a three year period|
|O||=||Openness measured by annual average of the sum of current payments and current receipts (goods, services, income and transfers) for a five year period|
|V||=||Variability of current receipts and net capital flows measured as standard deviation from centred three-year trend over a thirteen year period|
|R||=||Reserves measured by twelve month average over a year of official reserves (foreign exchange, SDR holdings, reserve position in the Fund and monetary gold)|
|k||=||Compression factor of 0.95 (applied to the uncompressed calculated quota shares which are then rescaled to sum to 100)|
The new quota formula represents a compromise. On the one hand, there is a desire for the quota calculation to follow the credo of simplicity.33 The fewer formulas and variables used, the less complex the calculation becomes. On the other hand, quota calculation has to meet the requirements of multifaceted roles and (potentially contradicting) functions of quotas.34 This requires more detailed calculations and may not be reflected by a simple formula.
If the new CQS is implemented without any changes, 115 developing and 15 developed countries will lose quota shares while 16 developed and 34 developing countries will gain quota shares and 5 countries' quotas will not change at all. Figure 2 shows the 30 countries that would experience the largest gains and losses (in percentage point difference) from a change to the newly calculated shares. China would benefit by far the most from the new quota formula with an increase in quota share of about 3.7 percentage points. At the other end of the spectrum, Saudi Arabia would lose approximately 2.4 percentage points. Overall, the G20 ambition to increase the quota shares of developing countries would be only partially met by implementing the new IMF quota formula. Thus another instrument for redistributing votes towards developing countries becomes important, namely the increase in the number of basic votes.
Figure 2 (back to the text)
30 Countries with Highest Gains and Losses by Change from Current Quota Share to New Calculated Quota Shares
Increase in Basic Votes
The Singapore reform package includes the tripling of basic votes for all members from the current 250 to 750 votes. The role of basic votes has always been debated between supporters of a UN-model (one country, one vote) and supporters of a bank model (votes according to economic power). On the one hand, a greater voice for developing countries would increase acceptance of IMF work and conditions.35 On the other hand, too large a role for developing countries could create a situation in which debtor countries would have more influence in the Fund than creditor countries. This could undermine developed countries' control, which is needed for their acceptance of their role as creditors36 but is sometimes considered as endangering the Fund's credibility in the international financial system.37 Between these two poles (represented by Eichengreen38 and Meltzer39 respectively), the G20 has committed itself to shift weight towards the first position, i.e. to improve voice and representation.
However, in practice, the role of basic votes has always been rather marginal. A reference point for this debate is the fact that basic votes accounted for 11% of total votes upon the IMF's founding in 1945. The maximal share of basic votes was about 15% in the 1970s, whereas the current share of approximately 2% is the lowest ever.40 The suggested tripling of basic votes would result in basic votes comprising a 6% share of total votes. Whether this is an ideal level or whether another level would be more appropriate is a political question.
Double Majority as Alternative
Another route toward increasing the impact of under-represented emerging and developing countries would affect the voting system. For example, the Stiglitz Commission41 advocates more double majorities, which would require not only a certain percentage of votes but also a certain share of members ("shares and chairs"42). Currently, a double majority (85% of votes and 60% of members) is only necessary for an amendment to the Fund's Articles. From the Stiglitz Commission's viewpoint, double majorities should be used for a broader range of decisions to give greater voice to small members. A drawback to the double majority principle is that it renders decision-making more difficult by imposing more restrictions. Therefore, small members would have greater influence only in the sense of more blocking power.
Interestingly, the Manuel Group43 and the IMF44 itself support the double majority principle and in addition call for lowering the 85% voting threshold to 70-75%. This would mean the abolishment of the present US de facto veto (with its voting share of approximately 17%). Hence, this decision would require an amendment of the Articles which itself would require the double majority of 85% of votes plus 60% of members; in other words, the United States would have to vote in favour of forgoing its own veto.
Figure 3 (back to the text)
Current, Calculated and New Calculated Quota Shares of Selected Member Countries in Relation to World GDP Share
Is Singapore Enough?
In any case, it seems highly questionable whether even full implementation of the new IMF quota formula will be sufficient to reach the stated objective, i.e. "to reflect changes in the world economy". As possible quantification for this objective, one may compare quotas to PPP-valued GDP. We do so by referring to three kinds of quotas. Figure 3 depicts the CQS that guided the decision of the last general quota review in 2006, the current quota distribution that has been effective since 2006, the new CQS based upon the Singapore quota formula suggested in 2008 for selected countries/regions, and these countries' weight as indicated by their relative share of world PPP-GDP in 2009.
Figure 3 shows marked differences between these four allocations of quota shares in the IMF. In some cases, clear trends become obvious: the European Union's share would decline as a result of quota reform, but it would still appear too high compared to its global share of PPP-GDP. By contrast, the United States is rather under-represented. There are some additional large countries that are obviously under-represented, including China, India, Mexico and Brazil. Japan seems to be properly represented. Africa is a more complicated case, as it is over-represented according to economic criteria. The same applies to Saudi Arabia, which received its high quota in the 1970s when the intention was that the IMF should benefit from Saudi Arabia's high revenues from oil sales. In summary, the new IMF quota formula (agreed on in Singapore) reflects the relative economic weight more properly, although it will hardly go far enough to give emerging and developing countries significant voice and representation.
The European Quota Problem
One of the biggest problems of present IMF quota discussion is the case of Europe. The 27 countries of the European Union presently have a quota share of about 32% and are thus the single most important power, assuming they can agree on their strategy. This situation can only be regarded as anachronistic over-representation, criticised even by influential European policymakers, for example by Bini Smaghi.45 To the emerging countries mentioned above, the EU's share must appear exaggerated, reflecting the world as it was after World War II rather than its present state. Even within the group of industrialised countries, the United States has roughly the same GDP and population but less than 60% of Europe's quota share. The European over-representation is also reflected in the distribution of seats within the important IMF Executive Board, where 24 executive directors basically agree unanimously on decisions.46 Due to this decision-making form, it is important to have representation here. Therefore, nearly all countries are organised in groups, so-called constituencies, which are represented by an executive director. The exceptions are the five most important countries, including France, Germany and the UK, which are not part of any constituencies but are instead each represented by their own executive director. In addition to these three European countries, there are another five executive directors from Europe. In summary, Europe controls one third of seats in the Executive Board, largely reflecting the European quota share (which is larger than the 32% quota share of the EU27).
Seen from the capital and quota shares, the European representation makes sense and one can even claim that Europe is under-represented when benchmarked by the current CQS (see Figure 3). Nevertheless, seen from the viewpoint of representation, as aimed for by the G20, the European control share at the IMF of one third is absurd. If one takes for example the United States as a role model in this respect, the appropriate European representation may be one seat plus, the plus reflecting the diversity of nations and organisations within Europe.
We are well aware that quotas have always been and will likely continue to be allocated according to a political process. Nevertheless, the credibility of a policy institution such as the IMF would profit from basing its decisions upon transparent criteria. Allocating voting shares derived from a certain amount of basic votes plus a quota derived from a reasonable formula would provide a prime example of such an orientation.
Seen from this point of view, it is disappointing that G20 leaders did not pledge the implementation of a reformed quota calculation system as the basis for the shift in representation. Moreover, it is unclear whether a 5% change is sufficient to level imbalances in representation. If we take the presently used CQS, for example, the percentage point differences of under-represented countries47 add up to 15% of the total quota share that would have to be realigned in order to reflect calculated quotas.
Nonetheless, history tells us that reaching a consensus on the Fund's quota distribution has always been very difficult. Colabella et al.48 point out that past total quota increases followed the principle of "equiproportionality": quotas were increased in total to reflect the need for augmented liquidity but were distributed in relation to previous quota shares, keeping the percentage distribution of quotas untouched. This is the main reason current quotas reflect neither the distribution nor the absolute sum of calculated quotas. Thus, planned quota alignment must also be viewed in relation to historical achievements, which dampen expectations. Against the background of this lowered expectation, one can assess a five per cent shift in quotas as a noteworthy agreement.
Figure 4 (back to the text)
Ratification of the Singapore Reform Package by IMF Member Countries and by G20 Countries
However, the bottom line of any reform is its implementation. What has become of the Singapore reform process that was confirmed by the G20? The outstanding Singapore reform elements can only be implemented by an amendment of the Fund's Articles49 and thus require a double majority, i.e. ratification by 60% of all members (i.e. 111 countries) that represent more than 85% of total votes. Therefore, the support of G20 members is important in order to advance ratification of the reform package. This is especially true considering that the Singapore reforms were meant to be a two-year reform process, i.e. they should already have been completed in 2008.50 However, it did not happen before the G20 met in London. What has happened since then?
More than a year has passed since the London Summit and the first statement pushing for the Singapore reforms in April 2009. But as of 29 January 2010, only 55 members have ratified the amendment,51 representing slightly more than a quarter of all members and 68% of total votes (see Figure 4). Included among these are only twelve G20 countries, meaning seven G20 countries have not yet ratified the resolution.52 Still, even if these countries were to ratify the reforms, this would only bring the total to 62 countries with almost 79% of total votes. Another 49 countries holding more than 6% of total votes would need to ratify the reform package before it could be fully implemented. If one compares these requirements to the current "momentum" in ratification cases, as shown by Figure 4, it is questionable whether the reforms would be implemented in a timely manner, even with G20 support. The more time that elapses and the more the economy recovers from the crisis, the less urgent the pushing of these reforms will be perceived.
Overall, the G20 decisions on governance reforms are modest compared to several earlier proposals,53 and most of the claims have already been raised for quite some time. Moreover, the decisions remain somewhat imprecise (e.g. the criterion of under-representation). In addition, several points that triggered heavy criticism remain untouched.54 For instance, the intended reforms do not address decision-making rules. Voting majority thresholds and the scope of double majorities would be left unchanged. The United States would maintain its veto power even if the new calculated quota shares were implemented one-to-one and basic votes were tripled. Another concern focuses on quota calculation, which may be streamlined and simplified but which - according to most proposals - would still be completely based on economic factors. By definition, this cannot ensure the representation of poor countries as aimed for by the G20.
Yet, the G20 decisions are a step in the right direction. While they underperform with respect to the expectations and claims that were expressed in some fields, they can be seen as a remarkable first step in comparison to historical achievements. However, they only deserve applause if they can be swiftly implemented. Otherwise, it will be difficult to claim any progress at all.