Report of the Commission of Experts of the President of the UN General Assembly on Reforms of the International Monetary and Financial System (This is an unauthorized and unedited interim draft whic
(This is an unauthorized and unedited interim draft which is being circulated as a
reference for the preparations for the UN Conference on the World Financial and
Economic Crisis and its Impact on Development (1-3 June 2009). The intention is to
solicit feedback and comments before the preparation and issuance of the final
report.)
Chapter 5: International Financial Innovations
1. In previous chapters, we have analyzed what macroeconomic policy and regulatory
reforms are needed to guarantee a sustainable and development friendly recovery of the
world economy. In chapter 4, we looked at reforms of current financial institutions and
broader institutional innovations. In this chapter we take a look at a final set of innovations:
those that relate to the global reserve system, the management of sovereign debt defaults,
and innovations aimed at better distributing the risks between lenders and borrowers in
world markets and at increasing development financing.
The Global Reserve System
2. Since the breakdown of the Bretton Woods system and the suspension of the gold
convertibility of the dollar, a system of flexible exchange rates among major currencies has
predominated. Although alternative national and regional currencies (such as the euro)
compete with each other as international reserve assets and means of international
settlement, the dollar has maintained its predominant role in both regards, a predominance
that became firmly established after the Second World War. In a significant sense, therefore,
the post-Bretton Woods system has been a “fiduciary dollar standard”.
3. This system has proven to be unstable, incompatible with global full employment, and
inequitable.
4. One of the main problems of the Bretton Woods system was identified by Robert Triffin
in the 1950s: the use of a national currency (the US dollar) as the international reserve currency
generated a difficult dilemma: dollar deficits were necessary to increase global liquidity, but at
the same time eroded the confidence in the dollar as a reserve currency, which generated in
particular increasing risks as to the capacity to maintain the dollar-gold parity. Although
abandonment of dollar convertibility and flexible exchanges rates eliminated some of these
problems, it created new ones. Instead of uncertainty over the ability to maintain the dollargold
parity, the “Triffin dilemma” is now associated with the large swings in the current
account imbalances of the U.S. and associated volatility of the dollar exchange rate, and in
the long run with the risk of the loss in the value of foreign exchange reserves held in dollars
as U.S. imbalances grow.
5. The instability and incapacity to guarantee full employment are also associated with the
fact that, even after the introduction of flexible exchange rates, the system was unable to
eliminate the deflationary (contractionary) bias associated with asymmetric adjustment to
payments imbalances falling on deficit countries –the fact that deficit countries face stronger
pressures to reduce their payments imbalances (the major exception being the reserve issuing
country) than surplus countries face to correct theirs. Flexible exchange rates also introduced
new forms of instability in a world of increasing free capital mobility: those associated with
the volatility of capital flows, particularly but not only short-term flows.
6. Finally, the inequities are generated by the fact that, as a result of a sequence of severe
crises, developing countries learnt that they need better instruments to protect themselves
against global financial and economic instability. Coupled with the increasing unwillingness
92
of developing countries to submit to the conditionality associated with IMF support, this has
led a massive accumulation of reserves over the past two decades. However, as these
reserves are mostly held in hard currency, they also represent a transfer of resources to the
United States and other industrialized countries.
7. Many believe that this problem could be eliminated by creating a supranational
international reserve currency. Indeed, the idea of an international reserve currency issued by
a supranational bank is not new. It was broached more than seventy five years ago by John
Maynard Keynes in his Treatise on Money, and refined in his Bretton Woods proposals for an
International Clearing Union. There currently exist a number of alternative proposals for a
new global reserve currency, for how the system might be administered, how the emissions
of the new currency might be allocated, and how the transition to the new system might be
managed. Considerable international discussion will be required for the international
community to decide the precise arrangements. However, this is an idea whose time has
come. This is a feasible proposal and it is imperative that the international community begins
working on the creation of such a new global reserve system. A failure to do so will
jeopardize prospects for a stable international financial system, which is necessary to support
a return to robust and stable growth.
Instability
8. The current international system is marred by a number of sources of instability. One of
the major problems, as noted, is that as holdings of the reserve currency accumulate over
time, confidence in its value as a store of value is likely to wane. After abandonment of fixed
exchange rates in the early 1970s, the main manifestation of the creation of excess dollar
liquidity was a tendency for the U.S. dollar to depreciate. When the U.S. responds with
action to reduce its external deficit–in part to restore the credibility of its reserve currency
status–this generates dollar appreciation accompanied by contractionary pressure on the
world economy. Irrespective of the phases of the global cycle of liquidity demand, U.S.
monetary policies are implemented with little consideration of their international impact and
are thus a potential cause of instability in exchange rates and global activity.
9. Since the 1960s, the system has indeed been plagued with cycles of confidence in the
U.S. dollar. These cycles have become particularly intense since the 1980s, leading to
unprecedented volatility both in the U.S. current account deficit and the effective exchange
rate of the U.S. dollar. As a result, the major attribute of an international store of value and
reserve asset, a stable external value, has been eroded. There is another sense in which the
current system is unstable.
10. By definition, for the world economy, the sum of all deficit countries’ balance of
payments must equal the sum of all other countries’ surpluses. But the way surplus and
deficits are brought into equality is not necessarily smooth and will usually involve changes
in incomes of individual countries. If a large number of countries choose policies aimed at
increasing their trade surpluses, or if international institutions encourage deficit countries to
improve their balance of payments, the deficits of the remaining country or countries will
become increasingly large. With the dollar as the major international reserve currency, unless
the U.S. is willing to be the “deficit country of last resort”, the adjustment will take place
through a decline in global income. In turn, if the U.S. macroeconomic policies are overtly
expansionary, unless other countries accept balance of payment surpluses the adjustment will
93
take place through expanding global income and inflation. In both cases, the result is likely
to be growing global imbalances, exchange rate instability and the erosion in confidence in
the dollar as a reserve currency.
11. The introduction of flexible exchange rates in the presence of growing private
international capital flows failed to meet the expectations that adjustment of balance of
payments would become smoother while leaving each country the necessary autonomy to
guarantee their domestic macroeconomic policy objectives. The basic reason is that
countries can avoid adjustment as long as they can attract sufficient external flows. When
these prove to be insufficient to cover funding for the imbalance or are reversed because of
lack of confidence, the adjustment takes the form of a financial crisis. The asymmetry
remains, but the negative impact on the deficit country is much greater, as the continued
financial crises since the mid-1970s has made clear.
Self-insurance and Deflationary Bias of the Global Reserve System
12. Global imbalances, associated in part to the way different countries reacted to the
financial instability of the late 1990s and early 2000s, played an important role in the
macroeconomic conditions leading to the current world financial crisis. The asymmetric
adjustments to these global imbalances in their turn played a part in generating the
insufficiency of global aggregate demand that has converted a U.S. financial disruption into a
global economic recession. Unless both problems are remedied, it will be difficult to restore
the economy to robust, stable growth.
13. These difficulties in the design of the international financial system led to large
accumulations of reserves by developing countries in recent years, and especially after the
Asian and Russian crises of 1997-1998. These crises, as those that preceded it in the late
1970s and early 1980s, showed that developing and emerging countries are subject to strong
pro-cyclical capital flows. If authorities react by allowing capital surges during booms to
generate rapid exchange rate appreciation and the build up of current account deficits, the
outcome is almost certainly a twin balance of payments and domestic financial crisis later on.
This problem is particularly acute when the boom is in the form of short-term, largely
speculative capital flows, a point that came to be increasingly recognized after the Asian
crisis. The decision to build stronger current account positions and to accumulate large
foreign exchange reserves in the face of booming capital inflows in 2004-2007 were
therefore a common response of these countries to create policy space to respond to the
negative impact of the expected recurrence of crises. Similarly, commodity exporting
countries have experienced in the past repeated crises, when sharp improvements in the
terms of trade are accompanied by unsustainable demand booms and by exchange rate
appreciation that generate “Dutch disease” effects. More generally, since the Asian crisis
developing and emerging market countries have found it increasingly attractive to save that
part of exceptional export proceeds that were considered to be temporary. High commodity
prices in the years preceding the current crisis exacerbated the problems that this posed for
global balances.
14. These policies could be considered as a form of “self-insurance” or “self-protection”
against capital reversals, adverse movements in the terms of trade and excessive exchange
rate volatility associated with financial crises. The fact that the only available “collective
94
insurance”, in the form of IMF financial assistance, is highly conditional and often imposes
procyclical policies during crises, reinforced the view that self-protection in the form of
reserve accumulation was a better strategy.3
15. As a result of these factors reserve accumulations rose to 11.7% of world GDP in 2007
vs. 5.6% a decade earlier, at the time the Asian crisis struck. Reserve accumulations in the
period 2003-2007, in the run up to the crisis, amounted to an annual average of $777 billion
a year or 1.6% of global GDP. The major concern is that if the current crisis is as long and
as deep as feared, and if the assistance provided to developing countries is inadequate, there
will be attempts to preserve strong external balances through protectionist measures, beggarthy-
neighbour exchange rate policies, and stronger “self-insurance” through reserve
accumulation–all measures that will impede a rapid response to the crisis.
16. When reserve accumulation is the result of current account surpluses and not only of the
attempt to offset private autonomous foreign capital inflows, there is a reduction in global
aggregate demand.4 In the past, the negative impact on global aggregate demand of these
reserve accumulations was offset by other countries running policies that resulted in large
current account deficits, particularly loose monetary and fiscal policies in the United States.
But such policies, as we have seen, have been the source of global instability.
17. The question posed by the autonomous reduction in United States’ deficits is, what will
now sustain global aggregate demand? It is unlikely to be another American bubble leading
to another period of large and unsustainable American deficits and the continuation of
global imbalances. Such a course risks a repeat of the current crisis. Thus, something has to
be done about the underlying sources of the insufficiency of global aggregate demand.
18. A global reserve currency whose creation was not linked to the external position of a
national economy could provide a better system to manage the deflationary bias that the
system faces during crisis, as well as the broader problems of instability analyzed above. It
would be possible to regulate the creation of global liquidity, and reduce the ability of a
reserve currency country to create excessive liquidity. And the system can be designed in
ways to put pressure on countries to reduce their surplus and thus reduce their contribution
to the insufficiency of global aggregate demand. This would contribute to the reduction of
global imbalances.
3 There may be other reasons, such as the need to provide for an aging population that would lead countries to
adopt policies to increase domestic savings and hold them in the form of foreign assets. The associated
“imbalances” would then simply reflect differences in the propensities of different countries to save and invest.
This has led to the general idea that financial flows would be from developed with high saving aging
populations to developing countries with younger populations and higher returns on investment. However, this
has not been verified in the statistics on international capital flows. Restrictions on the ability to use industrial
policies to encourage nascent industries in emerging countries (as many of the currently industrialized countries
did in their earlier phases of development) under recent WTO agreements may have led some countries to
substitute exchange rate policies to effect similar outcomes, and this too may have contributed to reserve
accumulations.
4 These reserves are sometimes called “owned reserves” to differentiate them from “borrowed reserves”, when
their counterpart are capital inflows.
95
19. Other innovations to improve risk sharing mechanisms would reduce the demand for
reserve accumulations, and therefore reduce the magnitude of the requisite emissions (see
below).
Inequities
20. The current system is also inequitable because it results in developing countries
transferring resources to the industrial countries that issue the reserve currencies. In
particular, the build up of dollar reserves represents lending to the United States at very low
interest rates. This transfer has increased through time due to the realization by developing
countries that large foreign exchange reserves are their only defence in a world of acute
financial and terms of trade instability.
21. Developing countries are, in effect, lending to developed countries large amounts at low
interest rates–in 2007, the last year for which data was available, $3.7 trillion. The difference
between the lending rate and the interest rate which these countries pay to developed
countries when they borrow from them is a transfer of resources to the reserve currency
countries that exceeds in value the foreign assistance that developing countries receive from
the developed countries. The fact that developing countries choose to hold such reserves
provides testimony to their perception of the costs of instability, of the adjustment costs that
they would have to bear if they did not have these reserves.
Cost to the reserve currency country
22. The United States also incurs costs associated with its role as supplying global reserves.
The demand for global reserves has led to increasing current account deficits in the United
States that have had adverse effects on U.S. domestic demand, but when dollars are held to
meet increased demands for liquidity in surplus countries, they fail to produce any
countervailing adjustment in foreign demand. In addition, periodic needs to correct these
deficits require contractionary monetary or fiscal policies that have adverse domestic effects
on the U.S. economy.
23. Countries holding substantial dollar reserves have started to call for a constraint on U.S.
policies that might cause depreciation in the international value of the dollar and thus a
decline in the value of their reserve holdings. China, as the major holder of dollar reserves,
has already noted the risks to its dollar reserves should the US adopt policies leading to a
depreciation of the dollar. The only way to respond to this call would mean for the U.S. a
loss of policy autonomy as it would have to take the effects on the rest of the world in
designing its monetary policy. Maintaining autonomy to U.S. policy, as it would be required
to respond the current crisis, would be the basic advantage for the U.S. to move to a global
reserve system, beyond the benefits it would receive from a more stable global financial and
economic system.
A two (or multiple) reserve currency system may be worse than a single reserve currency
24. It must be emphasized that a system based on multiple, competing reserve currencies
would not solve the inequities of the current system, as reserve assets would still be provided
by industrial countries. It would also add an additional element of instability to a purely
96
dollar-based system, associated with the exchange rate volatility among the currencies used
as reserve currencies. Indeed, this problem is already present in the current system. Such
volatility results in major gains and losses by central banks on their reserve holdings, a
feature that increases the risk associated with holding specific reserve assets and, therefore,
undermines their value as what they are meant to be: low-risk assets.
25. The basic advantage of a multi-polar reserve world is, of course, that it would provide
room for diversification. However, if central banks and private agents were to respond to
exchange rate fluctuations by changing the composition of their international assets, this
would feed into exchange rate instability. Under these conditions, a multiple currency reserve
system would generate growing calls for a fixed exchange rate arrangement, but fixing the
exchange rates among major currencies in a world of free capital mobility would be a
daunting task. It would also require policy coordination and loss of monetary sovereignty
that seems unlikely under current political conditions.
Call for a Global Reserve Currency
26. These long standing deficiencies in current arrangements have become manifest in the
lead up to the current global financial crisis, and can make it deeper. If countries choose
increased protection in the form of higher domestic saving and accumulation of reserves as a
response to the uncertainty of global market conditions, this would lead to further deepening
of the aggregate demand problems that the world economy is facing. The increases in the
U.S. national debt and the balance sheet of the U.S. Federal Reserve have led to concerns in
some quarters about the stability of the dollar as a store of value. The low (almost zero)
return on holdings of dollars means that those holding dollars as reserves are receiving
virtually no return in exchange for the foreign exchange rate risk which they bear.
27. These are among the reasons that it is desirable to adopt a truly global reserve currency.
Such a global reserve system can also reduce global risks since confidence in and stability of
the reserve currency would not depend on the vagaries of the economics and politics of a
single country.
28. The current crisis provides, in turn, an ideal opportunity to overcome the political
resistance to a new global monetary system. It has brought home problems posed by global
imbalances, international instability, and the current insufficiency of global aggregate
demand. A global reserve system is a critical step in addressing these problems, in ensuring
that as the global economy recovers it moves onto a path of strong growth without setting
the stage for another crisis in the future. It is also a propitious moment because the United
States may find its reserve currency status increasingly costly. Moreover, the US has
embarked on a response to the crisis that will involve large domestic and also potentially
large external imbalances, with unpredictable implications for the international reserve
system. Thus, both the United States and foreign exchange reserve holding countries may
actually find it acceptable to move in the direction of a new system. The former would be
able to take policy decisions with less concern about their global impact; the latter would be
less concerned about the impact of US policies on their reserve holdings.
Institutional frameworks for a new global reserve system
97
29. In setting up such a system, a number of details need to be worked out. Among these
are, who would issue the reserve currency, in what amounts, to whom would they be issued,
and under what conditions.
30. The issues are largely separable. The responsibility for managing the global reserve
system could be given to the IMF, which currently issues the only global currency, Special
Drawing Rights (SDRs), on which the system could be built, but it could also be given to a
new institution, such as a “Global Reserve Bank”. If we turn to existing institutions, this
should not inhibit asking more fundamental questions about the necessary reform of their
structure in support of the global monetary system.
31. In one possible approach, countries would agree to exchange their own currencies for
the new currency –say International Currency Certificates (ICC), which could be SDRs—and
vice-versa in much the same way as IMF quotas are made up today (except that developing
countries would only contribute their own national currencies, not the proportion of IMF
quotas in convertible currencies). This proposal would be equivalent to a system of
worldwide “swaps” among central banks. The global currency would thus be fully backed by
a basket of the currencies of all members.
32. In an alternative approach, the international agency in charge of creating global reserves
would simply issue the global currency, allocating ICC to the member countries, much as the
IMF Special Drawing Rights are issued today. There would be no “backing” for the global
currency, except the commitment of central banks to accept it in exchange for their own
currencies. This is what would give the ICC (or SDRs) the character of an international
reserve currency, the same way acceptance by citizens of payments in a national currency
gives it the character of the domestic money. However, if the issues of global currency
received by countries are considered deposits in the IMF or the Global Reserve Bank, and
the institution in charge of managing the system is allowed to buy the government bonds of
member countries or lend to them, then those investments would be the “backing” of the
global currency just as domestic moneys are “backed” today by the assets of national central
banks (the government bonds in their hands and their lending to private sector financial
institutions).
33. Under any of these schemes, countries could agree to hold a certain fraction of their
reserves in the global currency. The global reserve currency could also pay interest, at a rate
attractive enough to induce its use as an investment for central bank reserves. Exchange
rates would be managed according to the rules that each country chooses, subject to the
condition that exchange rate management does not affect other countries –a rule that is
already included in the IMF Articles of Agreement and must be subject to appropriate
surveillance. As with SDRs, the exchange rate of the global currency would be the weighted
average of a basket of convertible currencies, the composition of which would have to be
agreed.
34. In the alternative in which the global currency is considered to be a deposit in the IMF
or Global Reserve Bank, earnings by these institutions’ investments (lending to countries
undergoing balance of payments crises, or otherwise in Treasury securities of member
countries) would finance the interest paid to those countries that hold deposits of the global
currency (possibly in excess of the original issues they received). Obviously the major
98
advantage to holding the global currency is that the diversification away from individual
currencies would generate more stability in the value of reserve holdings.
35. The global currency could be allocated to countries on the basis of some formula
(“quota”), based on their weight in the world economy (GDP) or their needs (some
estimation of the demand for reserves). Since developing countries hold reserves which are,
in proportion to their GDP, several times those of industrial countries (26.4% of GDP in
2007 vs. 4.8% for high-income OECD countries), to manage the trade and capital account
volatility they face, a formula that would allocate the currency according to some definition
of demand for reserves would result in larger proportional allocations to these countries.
One possibility is, of course, to give developing countries all allocations. Note that the
current SDR allocation is based on a particular “quota” system, that of the IMF, which
continue to be subject to heated debate because richer countries on average get a larger share
of new allocations–i.e., the opposite to what a criteria based on need would indicate.
36. The allocation can and should have built into it incentives and/or penalties against
maintaining surpluses. Countries that maintained surpluses would lose all or part of their
quota allocation if they are not utilized in a timely manner to increase global demand.
37. The size of the annual emissions should be targeted to offset the increase in (nonborrowed)
reserves, i.e. reductions in global purchasing power resulting from reserve
accumulations. Simpler versions of this proposal would have annual emissions fixed at a
given rate of say $150 to $300 billion a year (the first figure corresponds to the world
demand for reserves in 1998-2002 but the demand for reserves was much larger in 2003-
2007, indicating that even $300 billion a year might be insufficient).
38. More sophisticated and elaborate versions of this proposal would have emissions
adjusted in a countercyclical way—larger emissions when global growth is below potential. It
might be easier to get global consensus on either of these simpler variants, but more detailed
versions would be able to support a variety of global needs (e.g. generate badly needed
revenues for development or global public goods).
39. One institutional way of establishing a new global reserve system is simply a broadening
of existing SDR arrangements, making their issuance automatic and regular. Doing so could
be viewed simply as completing the process that was begun in the 1960s, when SDRs were
created. The simplest version, as noted, is an annual issuance equivalent to the estimated
additional demand for foreign exchange reserves due to the growth of the world economy.
But they could be issued in a counter-cyclical fashion, therefore concentrating the issuance
during crisis periods. One of the advantages of using SDRs in such a counter-cyclical fashion
is that it would provide a mechanism for the IMF to play a more active role during crises.
40. Still another mechanism to manage SDRs in a counter-cyclical way was suggested by
IMF economist Jacques Polak three decades ago: providing all financing during crises with
SDR loans, which would generate emissions that would be automatically extinguished once
loans are paid back. This would be the global equivalent to what the central banks of
industrial countries have been doing on a massive scale during recent months.
41. Indeed, a large counter-cyclical issue of SDRs is the best mechanism to finance world
liquidity and official support to developing countries during the current crisis. This was
recognized by the G-20 in its decision to issue the equivalent of $250 billion in SDRs.
99
However, this decision also illustrates the problems associated to tying SDRs issuance to
IMF quotas, as somewhat less than $100 billion of the proposed emissions would benefit
developing countries. This implies that this issue is closely tied to the ongoing debate about
reform of IMF quotas. None of the proposed reforms to quotas deals adequately with the
issue of equity, and indicates that different rules may have to be applied to quotas and SDR
issues, as indicated above.
42. Although developing countries would only receive part of the allocations, the capacity of
the Fund to lend would be considerably enhanced if the current system was reformed in
such a way as unutilized SDRs, particularly from industrial countries, could be used by the
IMF to lend to member countries in need –such as the proposal of treating unused SDRs as
deposits in the IMF. However, unless there are strong reforms in the Fund’s practices, the
ability of the emissions to address the liquidity and macroeconomic management problems
noted earlier might be impaired, as developing countries might be reluctant to turn to the
IMF for funds. Reforms in that direction were adopted in March 2009 with the creation of
the Flexible Credit Line with only ex-ante conditionality, the doubling of all credit lines and
the elimination of structural benchmarks in conditional IMF lending. But additional reforms
to make access less onerous will be needed.
43. A simple way to further the use of SDR allocations to advance developmental objectives
(which may still require changing the Articles of Agreement) would be for the IMFC and the
IMF Board to allow the IMF to invest some of the funds made available through issuance of
SDRs in bonds issued by multilateral development banks. This would be similar to the
proposal for a “development link” made by the UNCTAD panel of experts in the 1960s (see
below).
44. Thus, a well designed global currency system would go a long way to correct the “Triffin
dilemma” and the tendency of the current system to generate large global imbalances and the
deflationary biases that are characteristic of balance of payments adjustments during crises.
Depending on the way emissions are allocated, the system could also correct the inequities
associated with the large demand for reserves by developing countries, and provide collective
insurance against future shocks. If emissions were issued in a countercyclical way, they could
perform an even more important role in stabilization.
Historical antecedents
45. When Keynes revised his idea of a global currency in his proposal for an International
Clearing Union, as part of the preparations for what became the Bretton Woods Conference,
his major concern was the elimination of asymmetric adjustment between deficit and surplus
countries leading to the tendency towards deficiency of global aggregate demand and a
constraint on the policy space needed for policies in support of full employment. He also
had in mind the significant payments imbalances that would characterize the post-war order
and therefore the need to provide a better source of liquidity, both globally and for countries
that would leave the war with structural payments deficits. Of course, the first of these
problems, the asymmetric adjustment, was not corrected by the Bretton Woods system, and
the second was only partly corrected.
100
46. In turn, when Special Drawing Rights (SDRs) were created in the 1960s, the major
concern was how to provide a more reliable source of global liquidity to gold and reserve
currency holdings (mainly dollars, but also British pound sterling at the time). It was believed
that the existing sources of international liquidity were not reliable, as they depended in the
first case on gold production and in the second on deficits of the reserve currency countries,
particularly the United States. As the initial problems of global liquidity –the “dollar
shortage”—were overcome, the attention shifted to risks of excessive dollar liquidity and,
particularly, that the U.S. gold reserves would not be sufficient to support dollar-gold
convertibility. This finally generated the demise of the Bretton Woods “dollar-gold-exchange
standard” and the adoption of flexible exchange rates among major currencies.
47. At the time SDRs were created, it was hoped that they would become a major
component of global reserves, thus creating a system in which the growth of global liquidity
would depend on deliberate international decisions. This expectation has not been fulfilled,
as SDRs have been created only episodically and in a total of a little over 20 billion SDRs,
which represent only a minimal fraction of current world reserves. The nature of the
problems of provision of global liquidity was obviously transformed with development of
the private financial markets in Eurodollars and other Euro currencies and the introduction
of a flexible exchange rate system. These problems associated with the provision of global
liquidity are less important today, except during extraordinary conjunctures such as those
generated during the severe shortage of liquidity, such as those created by the global liquidity
crisis in August 1998 and the world financial crisis since September 2008. But a major
problem remains: dependence of global liquidity on the vagrancies of US macroeconomic
policies and balance of payments imbalances, which can generate either excessive or limited
world liquidity. The recurrent problem of developing country access to international liquidity
is still an embedded feature of the system, as a result of the pro-cyclical capital flows.
Therefore, although there are no longer risks of insufficient liquidity in the international
system, there are problems associated with the control of global liquidity and significant
equity issues in the access to such liquidity by developing countries.
48. In Keynes’ initial proposal for a post-war arrangement, there was no need to address the
problem of equity in issuance since the creation of clearing credits was entirely endogenous.
This question was also evaded in the initial issuances of SDRs, although some ideas were
proposed at the time on how to tie the issuance of a global currency to development
financing, particularly in the proposal made by an UNCTAD expert panel to link the
question of liquidity provision for developed economies to the needs of developing
economies for development financing. But, as we have seen, equity issues cannot be ignored
today, as the current system subjects developing countries to recurrent problems of
illiquidity or induces them to accumulate large amounts of foreign exchange reserves.
Transition to the new system
49. The reform of the global reserve system could take place through a global agreement or
through more evolutionary approaches, including those that could build on a series of
regional initiatives.
50. If a large enough group of countries agreed to pool reserves in a system in which they
agreed to create and hold a common reserve currency, which they would stand ready to
exchange for their own currencies, a regional reserve system or even a system of near-global
101
coverage could be established without the agreement of all countries. So long as the new
currency is convertible into any hard currency that itself is convertible into other currencies,
it could serve effectively as a reserve currency. The countries participating might also agree
to reduce, over time, their holdings of other reserve currencies.
51. Membership in this new “Reserve Currency Association” could be open to all that
subscribe to its Articles of Agreement. The advantages of participation are sufficiently great
that it is likely to grow over time, embracing more countries, holding a greater fraction of
their reserves in the new global reserve currency. Eventually even the United States would
probably find it desirable to join. Thus, gradually, through a stable, evolutionary process, we
may have achieved the creation of a new Global Reserve System, an alternative to the
current system. Of course, there is also a risk of adverse selection: as long as participation is
voluntary, soft currency countries would be those more willing to participate, and
convertible global currencies outside the scheme could remain as the preferred currencies.
52. Existing regional agreements might provide an alternative way of evolving towards to a
global Reserve System. Regional mechanisms have the advantage of their own and can be
based either on swap arrangements among central banks or on foreign exchange reserve
pools. Given the reluctance of governments to give up control over their reserves, swap
arrangements may be more acceptable. Reserve pools offer, however, other advantages, such
as the possibility of allowing the reserve fund to borrow during periods of stress, and, as
noted, to issue a currency or reserve asset that could be used at a regional or global level. In
the 1980s, for example, the Latin American Reserve Fund was allowed to issue Andean
pesos.5 This asset, which has never been used, was expected to be used in intra-regional
trade, with periodic clearing of those held by central banks. The Chiang Mai Initiative,
created in 2000 by members of ASEAN, China, Japan and the Republic of Korea is another
important example of regional cooperation.6 Were this Initiative to evolve into a reserve
fund, it could back the issuance of a regional asset that could actually be attractive to central
banks in other parts of the world to hold as part of their reserve assets. However, if the
Chiang Mai Initiative is to achieve the objectives set forth for a global reserve currency, and
if it is to play a more effective role in stabilization, it would be necessary to eliminate the
provision that, after a certain threshold of use of existing swap facilities, countries would
have to be subject to IMF conditionality.
53. A common criticism of regional arrangements is that they are not effective in providing
diversification against systemic crisis, given that regional members are more likely to be
adversely affected at the same time. Although this implies that they are a complement and
not a substitute for a global solution, the underlying analysis is imprecise. Although the
ability of regional arrangements to address external shocks depends on negative events not
being correlated across participating countries, they could still be useful if shocks affect
member countries with different intensities or with lags, since this would allow some
5 The Latin American Reserve Fund was created by the Andean countries in the 1978 and was then called the
Andean Reserve Fund. Current members are Bolivia, Colombia, Costa Rica, Ecuador, Peru, Uruguay and
Venezuela.
6 This Initiative works as a system of bilateral swaps by member central banks, which are in the process of
being arranged on a multilateral basis. The system has so far not been used. ASEAN has a swap arrangement of
its own that has a longer history.
102
countries to lend their reserves to those experiencing a more severe or earlier shock.
Furthermore, lending at the onset of a liquidity squeeze could prevent a crisis in a given
country from affecting other countries, thereby reducing the correlation produced by
contagion. More generally, a country would benefits from the regional arrangement if the
variability of the regional reserve pool is lower than that of its individual reserves, and if
potential access to the pool reduces the possibility of attacks on individual members and
therefore acts as a mechanism of collective insurance that is more powerful than selfinsurance.
Statistical analysis by the UN Economic Commission for Latin American Reserve
and the Caribbean provides supports for this approach, as it indicates that correlations of
relevant macroeconomic variables among countries in the region may be lower than usually
assumed.
54. Regional initiatives could become an embedded part of the global reserve system. Some
have suggested that the reformed IMF should be a network of regional reserve funds. Such a
decentralized system would have many advantages, including the possibility of solving
problems associated with crises in the smaller countries at the regional level. The system
would also be attractive for medium and small-sized countries that could have a stronger
voice at the regional level. One way to link regional and global arrangements would be to
make contributions to regional arrangements one of the factors that could be taken into
account in determining SDR allocations.
SOVEREIGN DEBT DEFAULT AND RESTRUCTURING
Inadequacies of the existing system (or “non-system”) to manage debt crises
55. Sovereign debt crises have been a major source of the difficulties faced by developing
countries in achieving sustained growth and development at different times since the 1980s.
Social costs of these crises have been extremely large, and included long periods of lost
income and jobs, increased poverty and, in some cases, worsening income inequality. Given
the instability of external capital flows, severe financial crises hit even countries that were
judged by international opinion to have been soundly managed. In several cases, crises
originated in the need for a government to take over the responsibility for servicing privatesector
debts, of the banking system or key firms that were judged “too big to fail” –in a way
not too different to how the US and other industrial country governments have done during
in the current global crisis. In other cases, the inability of the central bank to provide foreign
exchange for private external debt servicing has led to effective official responsibility for the
debt. This “nationalization” of private sector external debts was indeed a feature of the Latin
American debt crisis of the 1980s and has been quite common in developing country debt
crises since then.
56. Not only are current “work-out” processes protracted and costly, often the debt writedown
has been insufficient to provide sustainability. The existence of debt overhangs
depresses growth, contributes to poverty, and crowds out essential public services. Often,
because write downs have been insufficient, they are soon followed by another crisis. And
because of the adverse terms and high costs imposed by the work-out, developing countries
are reluctant to default in a timely way, resulting in delays in dealing with the underlying
problems.
103
57. Moreover, worries about a protracted crisis in one country having spill-overs for others
has motivated massive bail-outs, contributing in turn to problems of moral hazard,
enhancing the likelihood of future crises.
58. Whether owing to risky policies or the intensified economic fluctuations of liberalized
financial environments, the existing system of protracted and creditor-biased resolution of
sovereign debt crises is not in the global public interest, and is far from the interest of the
poor in the affected countries.
59. The existing “system” (or really “non-system”) arose as piecemeal and mostly ad hoc
intergovernmental responses to sovereign debt crises as they occurred over the past halfcentury
or so. The fact that, as noted, the solutions that the current system provides take
time to be adopted and provide inadequate relief, implies that the system for addressing
sovereign debtors is clearly inferior to that provided in many countries for corporations and
sub-sovereign public entities by national bankruptcy regimes. The latter not only aim to find
a quick and equitable solution, but also one that achieves nationally desired economic and
social outcomes, particularly a ”fresh start” (or “clean slate”) when a bankrupt entity is
reorganized. The sovereign system is plagued also by horizontal inequities. Official lenders
have always complained that private creditors do not follow restructurings agreed in the
Paris Club, and the magnitude of debt rescheduling and relief accorded in individual cases
has clearly depended on the weight and negotiating capacity of the debtor country.
60. The system for sovereign debtors has operated under the informal and imperfect
coordination of the debtor and its creditors by the International Monetary Fund (IMF),
under the guidance of the G7 major industrialized countries. The latter countries set the
overall policy directions for the IMF and the other involved institutions, such as the Paris
Club, where debts owed to governments are restructured. The system assumes a developing
country government in debt distress will adopt an IMF-approved macroeconomic
adjustment program, that the program will be effective, and that all the relevant classes of
creditors (banks, bondholders and suppliers, government creditors and multilateral
institutions) will cooperate in providing the overall amount of relief and financial support
deemed necessary on the basis of the Fund documents. Often there is very little real debt
relief, only a mere rescheduling of the obligations. And often the magnitude of relief is based
on excessively optimistic growth projections–setting the stage for problems down the line.
61. These assumptions never fully held and what confidence there was in the system was
severely affected by how the Heavily Indebted Poor Countries (HIPC) Initiative and the
East Asian, Russian, Ecuadorian and Argentine crises were handled. The HIPC Initiative
was initially judged insufficient to give the poorest countries a fresh start and, after almost a
decade of long negotiations, it was supplemented in 2005 with the Multilateral Debt Relief
Initiative. Nevertheless, the HIPC initiative was represented the first comprehensive
approach to the solution of the debt problem of poor developing countries. The initiative
came along with a framework which placed poverty reduction strategies at the centre of
development cooperation, based in part on a broad social dialogue including the
participation of civil society.
62. Apart from that, some of the individual, non-HIPC renegotiations that took place after
the East Asian crisis are seen as unsatisfactory. Most country “workouts” from debt crises in
104
this period were under cooperative voluntary arrangements with the bondholders that did
not reduce the level of debt. The transparency of some of these renegotiations processes –
including the pressures exerted on debtor countries by other nations and IFIs–has also been
questioned.
63. Moreover, while creditors have a seat at the table, other claimants–government retirees,
for instance, which have been promised a particular level of pensions–do not. Chapter 9 of
the US bankruptcy code, which applies to municipalities and other sub-sovereign public
entities, gives priority to these “public” claimants on government revenues. In contrast,
international procedures seem to pay insufficient attention to their interests.
64. Finally, some critics of current practices suggest that they are unnecessarily “painful”
because they are designed to provide strong incentives for countries not to default on their
obligations. Small and weak countries may be forced to pay the price for ensuring that the
overall system exercises discipline on borrowers.
65. Argentine’s rapid growth after its 2001 default, in spite of the long delay in the final
resolution, shows that eliminating debt overhang can provide conditions for rapid economic
growth even in seemingly adverse conditions. Despite rapid growth, however, this country
faced significant problem regaining access to private financial markets.
Call for an International Debt Restructuring Court
66. Some have argued that new debt restructuring procedures are not needed; all that is
required are small reforms in debt contracts–e.g. collective action clauses. But no country
relies solely on collective action clauses for debt resolution, and there is no reason to believe
that doing so for international debt would be sufficient either. For instance, collective action
clauses do not provide effective means of resolving conflicts among classes of claimants.
67. It is easy to agree that the amount of debt relief accorded to different countries should
depend on their circumstances. However, it is artificial to have one set of rules for
determining that relief for selected developing countries–as was the case for the HIPCs and
later the Multilateral Debt Relief Initiative–and another for the rest of the world. Rather, we
need a single statutory framework for debt relief in which creditors and the debtor
restructure the debt so as to provide a fresh start, based on the country’s unique economic
condition. The debt workout regime should be efficient, equitable, transparent and timely in
handling debt problems ex post (as problems become apparent, and especially after default)
while promoting efficiency ex ante (when the borrowing takes place).
68. A well-designed process should protect the rights of minority, as well as majority,
creditors–as well as “public” claimants. It should give debtors the opportunity to call default
through a structured process. The principles of human-centred development, of
sustainability and of equity in the treatment of the debtor and its creditors, and among the
creditors, should apply equally to all sovereign debt crises resolved through the international
system. As in national bankruptcy systems, principals should be encouraged to reach a
workout on their own to the extent possible. But whether such an agreement can be reached,
and the nature of the agreement, can be affected by the backdrop of the legal structures.
105
69. Achieving these objectives require a more structured framework for international
cooperation in this area. For the same reason that governments adopt bankruptcy legislation
and do not rely solely on voluntary processes for resolving corporate bankruptcies, an
efficient sovereign system requires something more than a moral appeal to cooperation. This
means the creation of a sovereign debt workout mechanism.
70. This entails the creation of an “International Debt Restructuring Court”, similar to
national bankruptcy courts. This Court would ensure that agreed international principles
regarding priority of claims, necessary overall write downs and the sharing of “haircuts” were
followed. It could differentiate between distinct debt categories which might include
government, government guaranteed and government acquired private debt so as to make
transparent the actual effective liabilities of the sovereign. It could also determine what debts
could be considered “odious”. And it would be able to grant potential private or public
creditors authority to extend “debtor in possession” financing, as in corporate restructurings,
National courts would have to recognize the legitimacy of the international court, and both
creditors and debtors will therefore follow its rulings.
71. Once proceedings have started, the “Court” might act as a mediator, attempting to
establish international norms for sovereign debt restructurings. With a view to realizing a
comprehensive workout, it would encourage the creditors to coordinate their positions
within and across different classes of lenders, in the long run including the government
creditors that operate today through the Paris Club as well as multilateral creditors. Were
mediation to fail or become unduly lengthy, the Court should have the power to arbitrate. It
might also work in cooperation with the IMF, the World Bank, or regional development
banks, to help provide interim finance to maintain economic strength while negotiations take
place. But such lending should not be a mechanism simply for bailing out creditors who
failed to do due diligence in providing lending.
72. Beyond the problems of sovereign debt restructuring, there are also serious problems in
managing cross-border private debt workouts, with conflicts between different jurisdictions
and with concerns about “home” country bias. The International Debt Restructuring Court
could extend its reach to consider bankruptcy cases involving parties in multiple
jurisdictions.
73. In earlier discussions of sovereign debt restructuring mechanisms, it was presumed that
the IMF, or a separate and newly established division of the IMF, would act as the
bankruptcy court. However, while it may be desirable to institutionalize the sovereign debt
restructuring mechanism under the umbrella of an international institution, the IMF in its
current form is unlikely to be the appropriate institution, as it is also a creditor and subject to
disproportionate influence by creditor countries. It is therefore unlikely to be seen as a
“neutral” mediator. The arbitration process of the International Centre for the Settlement of
Investment Disputes (ICSID) within the World Bank has similarly failed to generate
confidence from the developing countries as a fair arbitrator of investor-state disputes under
bilateral investment agreements.
74. Any procedure must be based on widely shared principles and processes with political
legitimacy. Agreed upon goals–such as that the work-out must be fair, transparent
sustainable, and promote development–would boost its credibility with debtors–indeed with
106
all stakeholders, including creditors who would appreciate the reduction of uncertainty under
clear rules of the game and the knowledge that any post-workout debt situation would have
a larger chance of being sustainable. But translating these goals into agreed upon principles
and procedures may be difficult, given the conflicts of interests.
75. There is nothing immutable in the current approach to resolving sovereign debt crises. It
arose in the
reference for the preparations for the UN Conference on the World Financial and
Economic Crisis and its Impact on Development (1-3 June 2009). The intention is to
solicit feedback and comments before the preparation and issuance of the final
report.)
Chapter 5: International Financial Innovations
1. In previous chapters, we have analyzed what macroeconomic policy and regulatory
reforms are needed to guarantee a sustainable and development friendly recovery of the
world economy. In chapter 4, we looked at reforms of current financial institutions and
broader institutional innovations. In this chapter we take a look at a final set of innovations:
those that relate to the global reserve system, the management of sovereign debt defaults,
and innovations aimed at better distributing the risks between lenders and borrowers in
world markets and at increasing development financing.
The Global Reserve System
2. Since the breakdown of the Bretton Woods system and the suspension of the gold
convertibility of the dollar, a system of flexible exchange rates among major currencies has
predominated. Although alternative national and regional currencies (such as the euro)
compete with each other as international reserve assets and means of international
settlement, the dollar has maintained its predominant role in both regards, a predominance
that became firmly established after the Second World War. In a significant sense, therefore,
the post-Bretton Woods system has been a “fiduciary dollar standard”.
3. This system has proven to be unstable, incompatible with global full employment, and
inequitable.
4. One of the main problems of the Bretton Woods system was identified by Robert Triffin
in the 1950s: the use of a national currency (the US dollar) as the international reserve currency
generated a difficult dilemma: dollar deficits were necessary to increase global liquidity, but at
the same time eroded the confidence in the dollar as a reserve currency, which generated in
particular increasing risks as to the capacity to maintain the dollar-gold parity. Although
abandonment of dollar convertibility and flexible exchanges rates eliminated some of these
problems, it created new ones. Instead of uncertainty over the ability to maintain the dollargold
parity, the “Triffin dilemma” is now associated with the large swings in the current
account imbalances of the U.S. and associated volatility of the dollar exchange rate, and in
the long run with the risk of the loss in the value of foreign exchange reserves held in dollars
as U.S. imbalances grow.
5. The instability and incapacity to guarantee full employment are also associated with the
fact that, even after the introduction of flexible exchange rates, the system was unable to
eliminate the deflationary (contractionary) bias associated with asymmetric adjustment to
payments imbalances falling on deficit countries –the fact that deficit countries face stronger
pressures to reduce their payments imbalances (the major exception being the reserve issuing
country) than surplus countries face to correct theirs. Flexible exchange rates also introduced
new forms of instability in a world of increasing free capital mobility: those associated with
the volatility of capital flows, particularly but not only short-term flows.
6. Finally, the inequities are generated by the fact that, as a result of a sequence of severe
crises, developing countries learnt that they need better instruments to protect themselves
against global financial and economic instability. Coupled with the increasing unwillingness
92
of developing countries to submit to the conditionality associated with IMF support, this has
led a massive accumulation of reserves over the past two decades. However, as these
reserves are mostly held in hard currency, they also represent a transfer of resources to the
United States and other industrialized countries.
7. Many believe that this problem could be eliminated by creating a supranational
international reserve currency. Indeed, the idea of an international reserve currency issued by
a supranational bank is not new. It was broached more than seventy five years ago by John
Maynard Keynes in his Treatise on Money, and refined in his Bretton Woods proposals for an
International Clearing Union. There currently exist a number of alternative proposals for a
new global reserve currency, for how the system might be administered, how the emissions
of the new currency might be allocated, and how the transition to the new system might be
managed. Considerable international discussion will be required for the international
community to decide the precise arrangements. However, this is an idea whose time has
come. This is a feasible proposal and it is imperative that the international community begins
working on the creation of such a new global reserve system. A failure to do so will
jeopardize prospects for a stable international financial system, which is necessary to support
a return to robust and stable growth.
Instability
8. The current international system is marred by a number of sources of instability. One of
the major problems, as noted, is that as holdings of the reserve currency accumulate over
time, confidence in its value as a store of value is likely to wane. After abandonment of fixed
exchange rates in the early 1970s, the main manifestation of the creation of excess dollar
liquidity was a tendency for the U.S. dollar to depreciate. When the U.S. responds with
action to reduce its external deficit–in part to restore the credibility of its reserve currency
status–this generates dollar appreciation accompanied by contractionary pressure on the
world economy. Irrespective of the phases of the global cycle of liquidity demand, U.S.
monetary policies are implemented with little consideration of their international impact and
are thus a potential cause of instability in exchange rates and global activity.
9. Since the 1960s, the system has indeed been plagued with cycles of confidence in the
U.S. dollar. These cycles have become particularly intense since the 1980s, leading to
unprecedented volatility both in the U.S. current account deficit and the effective exchange
rate of the U.S. dollar. As a result, the major attribute of an international store of value and
reserve asset, a stable external value, has been eroded. There is another sense in which the
current system is unstable.
10. By definition, for the world economy, the sum of all deficit countries’ balance of
payments must equal the sum of all other countries’ surpluses. But the way surplus and
deficits are brought into equality is not necessarily smooth and will usually involve changes
in incomes of individual countries. If a large number of countries choose policies aimed at
increasing their trade surpluses, or if international institutions encourage deficit countries to
improve their balance of payments, the deficits of the remaining country or countries will
become increasingly large. With the dollar as the major international reserve currency, unless
the U.S. is willing to be the “deficit country of last resort”, the adjustment will take place
through a decline in global income. In turn, if the U.S. macroeconomic policies are overtly
expansionary, unless other countries accept balance of payment surpluses the adjustment will
93
take place through expanding global income and inflation. In both cases, the result is likely
to be growing global imbalances, exchange rate instability and the erosion in confidence in
the dollar as a reserve currency.
11. The introduction of flexible exchange rates in the presence of growing private
international capital flows failed to meet the expectations that adjustment of balance of
payments would become smoother while leaving each country the necessary autonomy to
guarantee their domestic macroeconomic policy objectives. The basic reason is that
countries can avoid adjustment as long as they can attract sufficient external flows. When
these prove to be insufficient to cover funding for the imbalance or are reversed because of
lack of confidence, the adjustment takes the form of a financial crisis. The asymmetry
remains, but the negative impact on the deficit country is much greater, as the continued
financial crises since the mid-1970s has made clear.
Self-insurance and Deflationary Bias of the Global Reserve System
12. Global imbalances, associated in part to the way different countries reacted to the
financial instability of the late 1990s and early 2000s, played an important role in the
macroeconomic conditions leading to the current world financial crisis. The asymmetric
adjustments to these global imbalances in their turn played a part in generating the
insufficiency of global aggregate demand that has converted a U.S. financial disruption into a
global economic recession. Unless both problems are remedied, it will be difficult to restore
the economy to robust, stable growth.
13. These difficulties in the design of the international financial system led to large
accumulations of reserves by developing countries in recent years, and especially after the
Asian and Russian crises of 1997-1998. These crises, as those that preceded it in the late
1970s and early 1980s, showed that developing and emerging countries are subject to strong
pro-cyclical capital flows. If authorities react by allowing capital surges during booms to
generate rapid exchange rate appreciation and the build up of current account deficits, the
outcome is almost certainly a twin balance of payments and domestic financial crisis later on.
This problem is particularly acute when the boom is in the form of short-term, largely
speculative capital flows, a point that came to be increasingly recognized after the Asian
crisis. The decision to build stronger current account positions and to accumulate large
foreign exchange reserves in the face of booming capital inflows in 2004-2007 were
therefore a common response of these countries to create policy space to respond to the
negative impact of the expected recurrence of crises. Similarly, commodity exporting
countries have experienced in the past repeated crises, when sharp improvements in the
terms of trade are accompanied by unsustainable demand booms and by exchange rate
appreciation that generate “Dutch disease” effects. More generally, since the Asian crisis
developing and emerging market countries have found it increasingly attractive to save that
part of exceptional export proceeds that were considered to be temporary. High commodity
prices in the years preceding the current crisis exacerbated the problems that this posed for
global balances.
14. These policies could be considered as a form of “self-insurance” or “self-protection”
against capital reversals, adverse movements in the terms of trade and excessive exchange
rate volatility associated with financial crises. The fact that the only available “collective
94
insurance”, in the form of IMF financial assistance, is highly conditional and often imposes
procyclical policies during crises, reinforced the view that self-protection in the form of
reserve accumulation was a better strategy.3
15. As a result of these factors reserve accumulations rose to 11.7% of world GDP in 2007
vs. 5.6% a decade earlier, at the time the Asian crisis struck. Reserve accumulations in the
period 2003-2007, in the run up to the crisis, amounted to an annual average of $777 billion
a year or 1.6% of global GDP. The major concern is that if the current crisis is as long and
as deep as feared, and if the assistance provided to developing countries is inadequate, there
will be attempts to preserve strong external balances through protectionist measures, beggarthy-
neighbour exchange rate policies, and stronger “self-insurance” through reserve
accumulation–all measures that will impede a rapid response to the crisis.
16. When reserve accumulation is the result of current account surpluses and not only of the
attempt to offset private autonomous foreign capital inflows, there is a reduction in global
aggregate demand.4 In the past, the negative impact on global aggregate demand of these
reserve accumulations was offset by other countries running policies that resulted in large
current account deficits, particularly loose monetary and fiscal policies in the United States.
But such policies, as we have seen, have been the source of global instability.
17. The question posed by the autonomous reduction in United States’ deficits is, what will
now sustain global aggregate demand? It is unlikely to be another American bubble leading
to another period of large and unsustainable American deficits and the continuation of
global imbalances. Such a course risks a repeat of the current crisis. Thus, something has to
be done about the underlying sources of the insufficiency of global aggregate demand.
18. A global reserve currency whose creation was not linked to the external position of a
national economy could provide a better system to manage the deflationary bias that the
system faces during crisis, as well as the broader problems of instability analyzed above. It
would be possible to regulate the creation of global liquidity, and reduce the ability of a
reserve currency country to create excessive liquidity. And the system can be designed in
ways to put pressure on countries to reduce their surplus and thus reduce their contribution
to the insufficiency of global aggregate demand. This would contribute to the reduction of
global imbalances.
3 There may be other reasons, such as the need to provide for an aging population that would lead countries to
adopt policies to increase domestic savings and hold them in the form of foreign assets. The associated
“imbalances” would then simply reflect differences in the propensities of different countries to save and invest.
This has led to the general idea that financial flows would be from developed with high saving aging
populations to developing countries with younger populations and higher returns on investment. However, this
has not been verified in the statistics on international capital flows. Restrictions on the ability to use industrial
policies to encourage nascent industries in emerging countries (as many of the currently industrialized countries
did in their earlier phases of development) under recent WTO agreements may have led some countries to
substitute exchange rate policies to effect similar outcomes, and this too may have contributed to reserve
accumulations.
4 These reserves are sometimes called “owned reserves” to differentiate them from “borrowed reserves”, when
their counterpart are capital inflows.
95
19. Other innovations to improve risk sharing mechanisms would reduce the demand for
reserve accumulations, and therefore reduce the magnitude of the requisite emissions (see
below).
Inequities
20. The current system is also inequitable because it results in developing countries
transferring resources to the industrial countries that issue the reserve currencies. In
particular, the build up of dollar reserves represents lending to the United States at very low
interest rates. This transfer has increased through time due to the realization by developing
countries that large foreign exchange reserves are their only defence in a world of acute
financial and terms of trade instability.
21. Developing countries are, in effect, lending to developed countries large amounts at low
interest rates–in 2007, the last year for which data was available, $3.7 trillion. The difference
between the lending rate and the interest rate which these countries pay to developed
countries when they borrow from them is a transfer of resources to the reserve currency
countries that exceeds in value the foreign assistance that developing countries receive from
the developed countries. The fact that developing countries choose to hold such reserves
provides testimony to their perception of the costs of instability, of the adjustment costs that
they would have to bear if they did not have these reserves.
Cost to the reserve currency country
22. The United States also incurs costs associated with its role as supplying global reserves.
The demand for global reserves has led to increasing current account deficits in the United
States that have had adverse effects on U.S. domestic demand, but when dollars are held to
meet increased demands for liquidity in surplus countries, they fail to produce any
countervailing adjustment in foreign demand. In addition, periodic needs to correct these
deficits require contractionary monetary or fiscal policies that have adverse domestic effects
on the U.S. economy.
23. Countries holding substantial dollar reserves have started to call for a constraint on U.S.
policies that might cause depreciation in the international value of the dollar and thus a
decline in the value of their reserve holdings. China, as the major holder of dollar reserves,
has already noted the risks to its dollar reserves should the US adopt policies leading to a
depreciation of the dollar. The only way to respond to this call would mean for the U.S. a
loss of policy autonomy as it would have to take the effects on the rest of the world in
designing its monetary policy. Maintaining autonomy to U.S. policy, as it would be required
to respond the current crisis, would be the basic advantage for the U.S. to move to a global
reserve system, beyond the benefits it would receive from a more stable global financial and
economic system.
A two (or multiple) reserve currency system may be worse than a single reserve currency
24. It must be emphasized that a system based on multiple, competing reserve currencies
would not solve the inequities of the current system, as reserve assets would still be provided
by industrial countries. It would also add an additional element of instability to a purely
96
dollar-based system, associated with the exchange rate volatility among the currencies used
as reserve currencies. Indeed, this problem is already present in the current system. Such
volatility results in major gains and losses by central banks on their reserve holdings, a
feature that increases the risk associated with holding specific reserve assets and, therefore,
undermines their value as what they are meant to be: low-risk assets.
25. The basic advantage of a multi-polar reserve world is, of course, that it would provide
room for diversification. However, if central banks and private agents were to respond to
exchange rate fluctuations by changing the composition of their international assets, this
would feed into exchange rate instability. Under these conditions, a multiple currency reserve
system would generate growing calls for a fixed exchange rate arrangement, but fixing the
exchange rates among major currencies in a world of free capital mobility would be a
daunting task. It would also require policy coordination and loss of monetary sovereignty
that seems unlikely under current political conditions.
Call for a Global Reserve Currency
26. These long standing deficiencies in current arrangements have become manifest in the
lead up to the current global financial crisis, and can make it deeper. If countries choose
increased protection in the form of higher domestic saving and accumulation of reserves as a
response to the uncertainty of global market conditions, this would lead to further deepening
of the aggregate demand problems that the world economy is facing. The increases in the
U.S. national debt and the balance sheet of the U.S. Federal Reserve have led to concerns in
some quarters about the stability of the dollar as a store of value. The low (almost zero)
return on holdings of dollars means that those holding dollars as reserves are receiving
virtually no return in exchange for the foreign exchange rate risk which they bear.
27. These are among the reasons that it is desirable to adopt a truly global reserve currency.
Such a global reserve system can also reduce global risks since confidence in and stability of
the reserve currency would not depend on the vagaries of the economics and politics of a
single country.
28. The current crisis provides, in turn, an ideal opportunity to overcome the political
resistance to a new global monetary system. It has brought home problems posed by global
imbalances, international instability, and the current insufficiency of global aggregate
demand. A global reserve system is a critical step in addressing these problems, in ensuring
that as the global economy recovers it moves onto a path of strong growth without setting
the stage for another crisis in the future. It is also a propitious moment because the United
States may find its reserve currency status increasingly costly. Moreover, the US has
embarked on a response to the crisis that will involve large domestic and also potentially
large external imbalances, with unpredictable implications for the international reserve
system. Thus, both the United States and foreign exchange reserve holding countries may
actually find it acceptable to move in the direction of a new system. The former would be
able to take policy decisions with less concern about their global impact; the latter would be
less concerned about the impact of US policies on their reserve holdings.
Institutional frameworks for a new global reserve system
97
29. In setting up such a system, a number of details need to be worked out. Among these
are, who would issue the reserve currency, in what amounts, to whom would they be issued,
and under what conditions.
30. The issues are largely separable. The responsibility for managing the global reserve
system could be given to the IMF, which currently issues the only global currency, Special
Drawing Rights (SDRs), on which the system could be built, but it could also be given to a
new institution, such as a “Global Reserve Bank”. If we turn to existing institutions, this
should not inhibit asking more fundamental questions about the necessary reform of their
structure in support of the global monetary system.
31. In one possible approach, countries would agree to exchange their own currencies for
the new currency –say International Currency Certificates (ICC), which could be SDRs—and
vice-versa in much the same way as IMF quotas are made up today (except that developing
countries would only contribute their own national currencies, not the proportion of IMF
quotas in convertible currencies). This proposal would be equivalent to a system of
worldwide “swaps” among central banks. The global currency would thus be fully backed by
a basket of the currencies of all members.
32. In an alternative approach, the international agency in charge of creating global reserves
would simply issue the global currency, allocating ICC to the member countries, much as the
IMF Special Drawing Rights are issued today. There would be no “backing” for the global
currency, except the commitment of central banks to accept it in exchange for their own
currencies. This is what would give the ICC (or SDRs) the character of an international
reserve currency, the same way acceptance by citizens of payments in a national currency
gives it the character of the domestic money. However, if the issues of global currency
received by countries are considered deposits in the IMF or the Global Reserve Bank, and
the institution in charge of managing the system is allowed to buy the government bonds of
member countries or lend to them, then those investments would be the “backing” of the
global currency just as domestic moneys are “backed” today by the assets of national central
banks (the government bonds in their hands and their lending to private sector financial
institutions).
33. Under any of these schemes, countries could agree to hold a certain fraction of their
reserves in the global currency. The global reserve currency could also pay interest, at a rate
attractive enough to induce its use as an investment for central bank reserves. Exchange
rates would be managed according to the rules that each country chooses, subject to the
condition that exchange rate management does not affect other countries –a rule that is
already included in the IMF Articles of Agreement and must be subject to appropriate
surveillance. As with SDRs, the exchange rate of the global currency would be the weighted
average of a basket of convertible currencies, the composition of which would have to be
agreed.
34. In the alternative in which the global currency is considered to be a deposit in the IMF
or Global Reserve Bank, earnings by these institutions’ investments (lending to countries
undergoing balance of payments crises, or otherwise in Treasury securities of member
countries) would finance the interest paid to those countries that hold deposits of the global
currency (possibly in excess of the original issues they received). Obviously the major
98
advantage to holding the global currency is that the diversification away from individual
currencies would generate more stability in the value of reserve holdings.
35. The global currency could be allocated to countries on the basis of some formula
(“quota”), based on their weight in the world economy (GDP) or their needs (some
estimation of the demand for reserves). Since developing countries hold reserves which are,
in proportion to their GDP, several times those of industrial countries (26.4% of GDP in
2007 vs. 4.8% for high-income OECD countries), to manage the trade and capital account
volatility they face, a formula that would allocate the currency according to some definition
of demand for reserves would result in larger proportional allocations to these countries.
One possibility is, of course, to give developing countries all allocations. Note that the
current SDR allocation is based on a particular “quota” system, that of the IMF, which
continue to be subject to heated debate because richer countries on average get a larger share
of new allocations–i.e., the opposite to what a criteria based on need would indicate.
36. The allocation can and should have built into it incentives and/or penalties against
maintaining surpluses. Countries that maintained surpluses would lose all or part of their
quota allocation if they are not utilized in a timely manner to increase global demand.
37. The size of the annual emissions should be targeted to offset the increase in (nonborrowed)
reserves, i.e. reductions in global purchasing power resulting from reserve
accumulations. Simpler versions of this proposal would have annual emissions fixed at a
given rate of say $150 to $300 billion a year (the first figure corresponds to the world
demand for reserves in 1998-2002 but the demand for reserves was much larger in 2003-
2007, indicating that even $300 billion a year might be insufficient).
38. More sophisticated and elaborate versions of this proposal would have emissions
adjusted in a countercyclical way—larger emissions when global growth is below potential. It
might be easier to get global consensus on either of these simpler variants, but more detailed
versions would be able to support a variety of global needs (e.g. generate badly needed
revenues for development or global public goods).
39. One institutional way of establishing a new global reserve system is simply a broadening
of existing SDR arrangements, making their issuance automatic and regular. Doing so could
be viewed simply as completing the process that was begun in the 1960s, when SDRs were
created. The simplest version, as noted, is an annual issuance equivalent to the estimated
additional demand for foreign exchange reserves due to the growth of the world economy.
But they could be issued in a counter-cyclical fashion, therefore concentrating the issuance
during crisis periods. One of the advantages of using SDRs in such a counter-cyclical fashion
is that it would provide a mechanism for the IMF to play a more active role during crises.
40. Still another mechanism to manage SDRs in a counter-cyclical way was suggested by
IMF economist Jacques Polak three decades ago: providing all financing during crises with
SDR loans, which would generate emissions that would be automatically extinguished once
loans are paid back. This would be the global equivalent to what the central banks of
industrial countries have been doing on a massive scale during recent months.
41. Indeed, a large counter-cyclical issue of SDRs is the best mechanism to finance world
liquidity and official support to developing countries during the current crisis. This was
recognized by the G-20 in its decision to issue the equivalent of $250 billion in SDRs.
99
However, this decision also illustrates the problems associated to tying SDRs issuance to
IMF quotas, as somewhat less than $100 billion of the proposed emissions would benefit
developing countries. This implies that this issue is closely tied to the ongoing debate about
reform of IMF quotas. None of the proposed reforms to quotas deals adequately with the
issue of equity, and indicates that different rules may have to be applied to quotas and SDR
issues, as indicated above.
42. Although developing countries would only receive part of the allocations, the capacity of
the Fund to lend would be considerably enhanced if the current system was reformed in
such a way as unutilized SDRs, particularly from industrial countries, could be used by the
IMF to lend to member countries in need –such as the proposal of treating unused SDRs as
deposits in the IMF. However, unless there are strong reforms in the Fund’s practices, the
ability of the emissions to address the liquidity and macroeconomic management problems
noted earlier might be impaired, as developing countries might be reluctant to turn to the
IMF for funds. Reforms in that direction were adopted in March 2009 with the creation of
the Flexible Credit Line with only ex-ante conditionality, the doubling of all credit lines and
the elimination of structural benchmarks in conditional IMF lending. But additional reforms
to make access less onerous will be needed.
43. A simple way to further the use of SDR allocations to advance developmental objectives
(which may still require changing the Articles of Agreement) would be for the IMFC and the
IMF Board to allow the IMF to invest some of the funds made available through issuance of
SDRs in bonds issued by multilateral development banks. This would be similar to the
proposal for a “development link” made by the UNCTAD panel of experts in the 1960s (see
below).
44. Thus, a well designed global currency system would go a long way to correct the “Triffin
dilemma” and the tendency of the current system to generate large global imbalances and the
deflationary biases that are characteristic of balance of payments adjustments during crises.
Depending on the way emissions are allocated, the system could also correct the inequities
associated with the large demand for reserves by developing countries, and provide collective
insurance against future shocks. If emissions were issued in a countercyclical way, they could
perform an even more important role in stabilization.
Historical antecedents
45. When Keynes revised his idea of a global currency in his proposal for an International
Clearing Union, as part of the preparations for what became the Bretton Woods Conference,
his major concern was the elimination of asymmetric adjustment between deficit and surplus
countries leading to the tendency towards deficiency of global aggregate demand and a
constraint on the policy space needed for policies in support of full employment. He also
had in mind the significant payments imbalances that would characterize the post-war order
and therefore the need to provide a better source of liquidity, both globally and for countries
that would leave the war with structural payments deficits. Of course, the first of these
problems, the asymmetric adjustment, was not corrected by the Bretton Woods system, and
the second was only partly corrected.
100
46. In turn, when Special Drawing Rights (SDRs) were created in the 1960s, the major
concern was how to provide a more reliable source of global liquidity to gold and reserve
currency holdings (mainly dollars, but also British pound sterling at the time). It was believed
that the existing sources of international liquidity were not reliable, as they depended in the
first case on gold production and in the second on deficits of the reserve currency countries,
particularly the United States. As the initial problems of global liquidity –the “dollar
shortage”—were overcome, the attention shifted to risks of excessive dollar liquidity and,
particularly, that the U.S. gold reserves would not be sufficient to support dollar-gold
convertibility. This finally generated the demise of the Bretton Woods “dollar-gold-exchange
standard” and the adoption of flexible exchange rates among major currencies.
47. At the time SDRs were created, it was hoped that they would become a major
component of global reserves, thus creating a system in which the growth of global liquidity
would depend on deliberate international decisions. This expectation has not been fulfilled,
as SDRs have been created only episodically and in a total of a little over 20 billion SDRs,
which represent only a minimal fraction of current world reserves. The nature of the
problems of provision of global liquidity was obviously transformed with development of
the private financial markets in Eurodollars and other Euro currencies and the introduction
of a flexible exchange rate system. These problems associated with the provision of global
liquidity are less important today, except during extraordinary conjunctures such as those
generated during the severe shortage of liquidity, such as those created by the global liquidity
crisis in August 1998 and the world financial crisis since September 2008. But a major
problem remains: dependence of global liquidity on the vagrancies of US macroeconomic
policies and balance of payments imbalances, which can generate either excessive or limited
world liquidity. The recurrent problem of developing country access to international liquidity
is still an embedded feature of the system, as a result of the pro-cyclical capital flows.
Therefore, although there are no longer risks of insufficient liquidity in the international
system, there are problems associated with the control of global liquidity and significant
equity issues in the access to such liquidity by developing countries.
48. In Keynes’ initial proposal for a post-war arrangement, there was no need to address the
problem of equity in issuance since the creation of clearing credits was entirely endogenous.
This question was also evaded in the initial issuances of SDRs, although some ideas were
proposed at the time on how to tie the issuance of a global currency to development
financing, particularly in the proposal made by an UNCTAD expert panel to link the
question of liquidity provision for developed economies to the needs of developing
economies for development financing. But, as we have seen, equity issues cannot be ignored
today, as the current system subjects developing countries to recurrent problems of
illiquidity or induces them to accumulate large amounts of foreign exchange reserves.
Transition to the new system
49. The reform of the global reserve system could take place through a global agreement or
through more evolutionary approaches, including those that could build on a series of
regional initiatives.
50. If a large enough group of countries agreed to pool reserves in a system in which they
agreed to create and hold a common reserve currency, which they would stand ready to
exchange for their own currencies, a regional reserve system or even a system of near-global
101
coverage could be established without the agreement of all countries. So long as the new
currency is convertible into any hard currency that itself is convertible into other currencies,
it could serve effectively as a reserve currency. The countries participating might also agree
to reduce, over time, their holdings of other reserve currencies.
51. Membership in this new “Reserve Currency Association” could be open to all that
subscribe to its Articles of Agreement. The advantages of participation are sufficiently great
that it is likely to grow over time, embracing more countries, holding a greater fraction of
their reserves in the new global reserve currency. Eventually even the United States would
probably find it desirable to join. Thus, gradually, through a stable, evolutionary process, we
may have achieved the creation of a new Global Reserve System, an alternative to the
current system. Of course, there is also a risk of adverse selection: as long as participation is
voluntary, soft currency countries would be those more willing to participate, and
convertible global currencies outside the scheme could remain as the preferred currencies.
52. Existing regional agreements might provide an alternative way of evolving towards to a
global Reserve System. Regional mechanisms have the advantage of their own and can be
based either on swap arrangements among central banks or on foreign exchange reserve
pools. Given the reluctance of governments to give up control over their reserves, swap
arrangements may be more acceptable. Reserve pools offer, however, other advantages, such
as the possibility of allowing the reserve fund to borrow during periods of stress, and, as
noted, to issue a currency or reserve asset that could be used at a regional or global level. In
the 1980s, for example, the Latin American Reserve Fund was allowed to issue Andean
pesos.5 This asset, which has never been used, was expected to be used in intra-regional
trade, with periodic clearing of those held by central banks. The Chiang Mai Initiative,
created in 2000 by members of ASEAN, China, Japan and the Republic of Korea is another
important example of regional cooperation.6 Were this Initiative to evolve into a reserve
fund, it could back the issuance of a regional asset that could actually be attractive to central
banks in other parts of the world to hold as part of their reserve assets. However, if the
Chiang Mai Initiative is to achieve the objectives set forth for a global reserve currency, and
if it is to play a more effective role in stabilization, it would be necessary to eliminate the
provision that, after a certain threshold of use of existing swap facilities, countries would
have to be subject to IMF conditionality.
53. A common criticism of regional arrangements is that they are not effective in providing
diversification against systemic crisis, given that regional members are more likely to be
adversely affected at the same time. Although this implies that they are a complement and
not a substitute for a global solution, the underlying analysis is imprecise. Although the
ability of regional arrangements to address external shocks depends on negative events not
being correlated across participating countries, they could still be useful if shocks affect
member countries with different intensities or with lags, since this would allow some
5 The Latin American Reserve Fund was created by the Andean countries in the 1978 and was then called the
Andean Reserve Fund. Current members are Bolivia, Colombia, Costa Rica, Ecuador, Peru, Uruguay and
Venezuela.
6 This Initiative works as a system of bilateral swaps by member central banks, which are in the process of
being arranged on a multilateral basis. The system has so far not been used. ASEAN has a swap arrangement of
its own that has a longer history.
102
countries to lend their reserves to those experiencing a more severe or earlier shock.
Furthermore, lending at the onset of a liquidity squeeze could prevent a crisis in a given
country from affecting other countries, thereby reducing the correlation produced by
contagion. More generally, a country would benefits from the regional arrangement if the
variability of the regional reserve pool is lower than that of its individual reserves, and if
potential access to the pool reduces the possibility of attacks on individual members and
therefore acts as a mechanism of collective insurance that is more powerful than selfinsurance.
Statistical analysis by the UN Economic Commission for Latin American Reserve
and the Caribbean provides supports for this approach, as it indicates that correlations of
relevant macroeconomic variables among countries in the region may be lower than usually
assumed.
54. Regional initiatives could become an embedded part of the global reserve system. Some
have suggested that the reformed IMF should be a network of regional reserve funds. Such a
decentralized system would have many advantages, including the possibility of solving
problems associated with crises in the smaller countries at the regional level. The system
would also be attractive for medium and small-sized countries that could have a stronger
voice at the regional level. One way to link regional and global arrangements would be to
make contributions to regional arrangements one of the factors that could be taken into
account in determining SDR allocations.
SOVEREIGN DEBT DEFAULT AND RESTRUCTURING
Inadequacies of the existing system (or “non-system”) to manage debt crises
55. Sovereign debt crises have been a major source of the difficulties faced by developing
countries in achieving sustained growth and development at different times since the 1980s.
Social costs of these crises have been extremely large, and included long periods of lost
income and jobs, increased poverty and, in some cases, worsening income inequality. Given
the instability of external capital flows, severe financial crises hit even countries that were
judged by international opinion to have been soundly managed. In several cases, crises
originated in the need for a government to take over the responsibility for servicing privatesector
debts, of the banking system or key firms that were judged “too big to fail” –in a way
not too different to how the US and other industrial country governments have done during
in the current global crisis. In other cases, the inability of the central bank to provide foreign
exchange for private external debt servicing has led to effective official responsibility for the
debt. This “nationalization” of private sector external debts was indeed a feature of the Latin
American debt crisis of the 1980s and has been quite common in developing country debt
crises since then.
56. Not only are current “work-out” processes protracted and costly, often the debt writedown
has been insufficient to provide sustainability. The existence of debt overhangs
depresses growth, contributes to poverty, and crowds out essential public services. Often,
because write downs have been insufficient, they are soon followed by another crisis. And
because of the adverse terms and high costs imposed by the work-out, developing countries
are reluctant to default in a timely way, resulting in delays in dealing with the underlying
problems.
103
57. Moreover, worries about a protracted crisis in one country having spill-overs for others
has motivated massive bail-outs, contributing in turn to problems of moral hazard,
enhancing the likelihood of future crises.
58. Whether owing to risky policies or the intensified economic fluctuations of liberalized
financial environments, the existing system of protracted and creditor-biased resolution of
sovereign debt crises is not in the global public interest, and is far from the interest of the
poor in the affected countries.
59. The existing “system” (or really “non-system”) arose as piecemeal and mostly ad hoc
intergovernmental responses to sovereign debt crises as they occurred over the past halfcentury
or so. The fact that, as noted, the solutions that the current system provides take
time to be adopted and provide inadequate relief, implies that the system for addressing
sovereign debtors is clearly inferior to that provided in many countries for corporations and
sub-sovereign public entities by national bankruptcy regimes. The latter not only aim to find
a quick and equitable solution, but also one that achieves nationally desired economic and
social outcomes, particularly a ”fresh start” (or “clean slate”) when a bankrupt entity is
reorganized. The sovereign system is plagued also by horizontal inequities. Official lenders
have always complained that private creditors do not follow restructurings agreed in the
Paris Club, and the magnitude of debt rescheduling and relief accorded in individual cases
has clearly depended on the weight and negotiating capacity of the debtor country.
60. The system for sovereign debtors has operated under the informal and imperfect
coordination of the debtor and its creditors by the International Monetary Fund (IMF),
under the guidance of the G7 major industrialized countries. The latter countries set the
overall policy directions for the IMF and the other involved institutions, such as the Paris
Club, where debts owed to governments are restructured. The system assumes a developing
country government in debt distress will adopt an IMF-approved macroeconomic
adjustment program, that the program will be effective, and that all the relevant classes of
creditors (banks, bondholders and suppliers, government creditors and multilateral
institutions) will cooperate in providing the overall amount of relief and financial support
deemed necessary on the basis of the Fund documents. Often there is very little real debt
relief, only a mere rescheduling of the obligations. And often the magnitude of relief is based
on excessively optimistic growth projections–setting the stage for problems down the line.
61. These assumptions never fully held and what confidence there was in the system was
severely affected by how the Heavily Indebted Poor Countries (HIPC) Initiative and the
East Asian, Russian, Ecuadorian and Argentine crises were handled. The HIPC Initiative
was initially judged insufficient to give the poorest countries a fresh start and, after almost a
decade of long negotiations, it was supplemented in 2005 with the Multilateral Debt Relief
Initiative. Nevertheless, the HIPC initiative was represented the first comprehensive
approach to the solution of the debt problem of poor developing countries. The initiative
came along with a framework which placed poverty reduction strategies at the centre of
development cooperation, based in part on a broad social dialogue including the
participation of civil society.
62. Apart from that, some of the individual, non-HIPC renegotiations that took place after
the East Asian crisis are seen as unsatisfactory. Most country “workouts” from debt crises in
104
this period were under cooperative voluntary arrangements with the bondholders that did
not reduce the level of debt. The transparency of some of these renegotiations processes –
including the pressures exerted on debtor countries by other nations and IFIs–has also been
questioned.
63. Moreover, while creditors have a seat at the table, other claimants–government retirees,
for instance, which have been promised a particular level of pensions–do not. Chapter 9 of
the US bankruptcy code, which applies to municipalities and other sub-sovereign public
entities, gives priority to these “public” claimants on government revenues. In contrast,
international procedures seem to pay insufficient attention to their interests.
64. Finally, some critics of current practices suggest that they are unnecessarily “painful”
because they are designed to provide strong incentives for countries not to default on their
obligations. Small and weak countries may be forced to pay the price for ensuring that the
overall system exercises discipline on borrowers.
65. Argentine’s rapid growth after its 2001 default, in spite of the long delay in the final
resolution, shows that eliminating debt overhang can provide conditions for rapid economic
growth even in seemingly adverse conditions. Despite rapid growth, however, this country
faced significant problem regaining access to private financial markets.
Call for an International Debt Restructuring Court
66. Some have argued that new debt restructuring procedures are not needed; all that is
required are small reforms in debt contracts–e.g. collective action clauses. But no country
relies solely on collective action clauses for debt resolution, and there is no reason to believe
that doing so for international debt would be sufficient either. For instance, collective action
clauses do not provide effective means of resolving conflicts among classes of claimants.
67. It is easy to agree that the amount of debt relief accorded to different countries should
depend on their circumstances. However, it is artificial to have one set of rules for
determining that relief for selected developing countries–as was the case for the HIPCs and
later the Multilateral Debt Relief Initiative–and another for the rest of the world. Rather, we
need a single statutory framework for debt relief in which creditors and the debtor
restructure the debt so as to provide a fresh start, based on the country’s unique economic
condition. The debt workout regime should be efficient, equitable, transparent and timely in
handling debt problems ex post (as problems become apparent, and especially after default)
while promoting efficiency ex ante (when the borrowing takes place).
68. A well-designed process should protect the rights of minority, as well as majority,
creditors–as well as “public” claimants. It should give debtors the opportunity to call default
through a structured process. The principles of human-centred development, of
sustainability and of equity in the treatment of the debtor and its creditors, and among the
creditors, should apply equally to all sovereign debt crises resolved through the international
system. As in national bankruptcy systems, principals should be encouraged to reach a
workout on their own to the extent possible. But whether such an agreement can be reached,
and the nature of the agreement, can be affected by the backdrop of the legal structures.
105
69. Achieving these objectives require a more structured framework for international
cooperation in this area. For the same reason that governments adopt bankruptcy legislation
and do not rely solely on voluntary processes for resolving corporate bankruptcies, an
efficient sovereign system requires something more than a moral appeal to cooperation. This
means the creation of a sovereign debt workout mechanism.
70. This entails the creation of an “International Debt Restructuring Court”, similar to
national bankruptcy courts. This Court would ensure that agreed international principles
regarding priority of claims, necessary overall write downs and the sharing of “haircuts” were
followed. It could differentiate between distinct debt categories which might include
government, government guaranteed and government acquired private debt so as to make
transparent the actual effective liabilities of the sovereign. It could also determine what debts
could be considered “odious”. And it would be able to grant potential private or public
creditors authority to extend “debtor in possession” financing, as in corporate restructurings,
National courts would have to recognize the legitimacy of the international court, and both
creditors and debtors will therefore follow its rulings.
71. Once proceedings have started, the “Court” might act as a mediator, attempting to
establish international norms for sovereign debt restructurings. With a view to realizing a
comprehensive workout, it would encourage the creditors to coordinate their positions
within and across different classes of lenders, in the long run including the government
creditors that operate today through the Paris Club as well as multilateral creditors. Were
mediation to fail or become unduly lengthy, the Court should have the power to arbitrate. It
might also work in cooperation with the IMF, the World Bank, or regional development
banks, to help provide interim finance to maintain economic strength while negotiations take
place. But such lending should not be a mechanism simply for bailing out creditors who
failed to do due diligence in providing lending.
72. Beyond the problems of sovereign debt restructuring, there are also serious problems in
managing cross-border private debt workouts, with conflicts between different jurisdictions
and with concerns about “home” country bias. The International Debt Restructuring Court
could extend its reach to consider bankruptcy cases involving parties in multiple
jurisdictions.
73. In earlier discussions of sovereign debt restructuring mechanisms, it was presumed that
the IMF, or a separate and newly established division of the IMF, would act as the
bankruptcy court. However, while it may be desirable to institutionalize the sovereign debt
restructuring mechanism under the umbrella of an international institution, the IMF in its
current form is unlikely to be the appropriate institution, as it is also a creditor and subject to
disproportionate influence by creditor countries. It is therefore unlikely to be seen as a
“neutral” mediator. The arbitration process of the International Centre for the Settlement of
Investment Disputes (ICSID) within the World Bank has similarly failed to generate
confidence from the developing countries as a fair arbitrator of investor-state disputes under
bilateral investment agreements.
74. Any procedure must be based on widely shared principles and processes with political
legitimacy. Agreed upon goals–such as that the work-out must be fair, transparent
sustainable, and promote development–would boost its credibility with debtors–indeed with
106
all stakeholders, including creditors who would appreciate the reduction of uncertainty under
clear rules of the game and the knowledge that any post-workout debt situation would have
a larger chance of being sustainable. But translating these goals into agreed upon principles
and procedures may be difficult, given the conflicts of interests.
75. There is nothing immutable in the current approach to resolving sovereign debt crises. It
arose in the
political and economic environment created after the Second World War, and
the need to develop a better system remains on the international policy agenda. The
international community needs to actively resume the effort to define the specific
mechanism to institutionalize the principles advanced here.
Foreign Debt Commission
76. The crisis also gives urgency to reform of institutional structures for debt relief, as an
increasing number of developing countries, especially the most vulnerable low-income
countries, may face difficulties in meeting their external debt commitments. This crisis
therefore gives urgency to these reforms. Unless these debts are managed better than they
have been in the past, the consequences for developing countries, and especially the poor in
these countries, can be serious.
77. Although, as argued above, there is a need for new procedures for restructuring
sovereign debt, this reform will take time, as it would require a new international treaty. In
the interim, something needs to be done to ensure that debts are better managed–and this
may be true even in the long run. It is important to take actions to manage debt better, so
that countries are not forced into default.
78. The United Nations should therefore set up a Foreign Debt Commission to consider
external debt problems of developing countries and economies in transition. The
commission, with balanced geographic representation and technical support from the
Bretton Woods, regional and other financial institutions, would address these issues and
provide advice on ways to enhance external debt crisis prevention and resolution.7 It would
also examine existing and advise on the design of better debt sustainability frameworks for
the international community to follow. It would help debt-distressed countries return to debt
sustainability, extend Paris Club-plus type approaches to new official creditors, set up an
interim mediation service, and craft on the basis of that experience more permanent debt
mediation and arbitration mechanisms.
Innovative Risk Management Instruments
79. The volatility of private capital flows to developing countries has generated increasing
demand for policies and instruments that would allow these countries to better manage and
minimize the risks associated with increasing international financial integration and, in
particular, to better distribute the risks associated with this integration among different
market agents. As has been demonstrated during past and current crises, the pro-cyclical and
7 See United Nations, “Doha Declaration on Financing for Development: outcome document of the Follow-up
International Conference on Financing for Development to Review the Implementation of the Monterrey
Consensus” (A/CONF.212/L.1/Rev.1), Doha, Qatar, 29 November -2 December 2008, paragraph 67.
107
herding behaviour of international capital flows tends to generate boom-bust cycles, which
are particularly damaging for developing countries, and it also reduces the scope they have to
undertake counter-cyclical macroeconomic policies. Moreover, many developing and
emerging countries borrow short term, in hard currencies, forcing them to bear the risk of
interest rate and exchange rate fluctuations. And finally, inadequate debt resolution
mechanisms impose high costs on developing countries.
80. In light of this, there have been a variety of ideas and proposals for introduction of
innovative financial instruments. The proposed instruments include tools that enable better
management of risks arising from the business cycle and fluctuations in commodity prices,
particularly GDP and commodity linked bonds and financial guarantees that have a countercyclical
element embedded in their structure. Promoting local currency bond markets has
also been seen as a way to enhance financial development and reduce the currency
mismatches that affect debt structures in developing countries.
81. GDP-linked bonds are conventional bonds that pay a low fixed coupon augmented by
an additional payment linked by a pre-determined formula to debtor country’s GDP growth.
This variable return structure links returns to the ability to service and thus reduces the
likelihood of costly and disruptive defaults and debt crises. The reduction of a country’s debt
service when the economy faces financing difficulties can also facilitate a more rapid
recovery, as it allows higher public spending in difficult times. For investors GDP-linked
bonds reduce the probability of default, and thus the costs of expensive renegotiation, and
offer a valuable diversification opportunity. Returns should be higher than with conventional
bonds.8
82. Since private financial markets are unlikely to develop these instruments autonomously,
because of the externalities associated with their introduction multilateral development banks
should take an active role in their development. In particular, these institutions could have
an active role as “market-makers”. The expertise developed by the World Bank as marketmaker
for the sale of carbon credits under the Kyoto protocol provides a precedent for these
activities. The World Bank and regional development banks could, for example, make loans
whose servicing would be linked to GDP. The loans could then be sold to financial markets,
either individually or grouped and securitized. Alternatively the World Bank or regional
banks could buy GDP-linked bonds that developing countries would issue via private
placements. The fact that major multilateral development banks became active in this type of
lending could, furthermore extend the benefits of adjusting debt service to growth variations
to countries that do not have access to the private bond market. GDP-indexed securities are
particularly appropriate for Islamic finance, as they can be made compatible with Sharia law
which prohibits charging of interest.
8 The introduction of these securities must overcome, however, some practical difficulties. One possible set of
concerns is associated with lags in the provision and frequent revisions of GDP data, as well as over the quality
of these estimates, but these issues should be easy to resolve through international standard setting and the
provision of technical assistance. More important in this regard is how to manage concerns that have been
raised about the liquidity of such instruments, especially when they are newly issued. Such concerns were
similarly raised when inflation indexed bonds were first introduced, but they are now accepted worldwide.
Governments and multilaterals might have to help create a thicker market.
108
83. There might also be alternative ways of ensuring flexible payment arrangements that
would allow automatic adjustment for borrowers during bad times. For instance, one
possibility is for coupon payments to remain fixed and the amortization schedule to be
adjusted instead. Countries would postpone part or all of their debt payments during
economic downturns; and they would then make up by pre-paying during economic
upswings. A historical precedent was set by the United Kingdom when it borrowed from the
United States in the 1940s. The Anglo American Financial Agreement was negotiated by
John Maynard Keynes and included a “bisque clause” that provided a waiver of 2% interest
payment in any year in which the United Kingdom foreign exchange income was not
sufficient to meet its pre war level of imports, adjusted to current prices.
84. Commodity-linked bonds can also play a useful role in reducing country vulnerabilities.
Examples of commodity-indexed bonds include oil-backed bonds, such as the Brady bonds
with oil warrants that were first issued on behalf of the government of Mexico. In such
instruments, the coupon or principal payments are linked to the price of a referenced
commodity. Again, it might be desirable for international institutions to help create a market
for such bonds.
85. However, the greater complexity of this instrument, in comparison with conventional
bonds, and the commodity price risk that the investor faces, may make commodity-linked
bonds expensive. Again, it might be desirable for international institutions to help create a
market for such bonds. While they are likely to be less useful than GDP-indexed bonds for
the growing number of developing countries that have a fairly diversified export structure
and therefore lack a natural commodity price to link to bond payments, they have the
decided advantage that the risk being “insured” through the bond is not one affected by the
actions of the country (i.e. moral hazard is less of a problem.)
86. Another way of addressing the problems created by the inherent tendency of private
flows to be pro-cyclical is for public institutions to issue guarantees that have countercyclical
elements. For example, Multilateral Development Banks (MDBs) and Export Credit
Agencies (ECAs) could introduce an explicit counter-cyclical element in all the risk
evaluations and the guarantees they issue for lending to developing countries. This requires
MDBs and ECAs to assess risk for issuing guarantees with a long-term perspective. When
banks or other lenders lowered their exposure to a country, MDBs or ECAs would increase
their level of guarantees, if they considered that the country’s long-term fundamentals were
basically sound. When matters were seen by private banks to improve, and their willingness
to lend increased, MDBs or ECAs could reduce their exposure. Alternatively, there could be
special stand-alone guarantee mechanisms for trade and/or long-term credit, for example
within multilateral or regional development banks, which had a strong explicit countercyclical
element. This could be activated in periods of sharp decline in capital flows and its
aim would be to try to catalyze private sector trade or long-term credits, especially for
infrastructure.
87. Finally, a number of developing countries have encouraged development of domestic
capital markets, and in particular local currency bond markets. These markets in fact
boomed after the Asian crisis, multiplying fivefold between 1997 and 2007 for the twenty
large and medium-sized emerging economies for which the Bank of International
Settlements provide regular information. This trend can be seen as a response of emerging
109
economies to the volatility and pro-cyclical bias of international capital flows and the
volatility of exchange rates. It can be viewed as a means of creating a more stable source of
local currency funding for both the public and private sectors, thereby mitigating the funding
difficulties created by sudden stops in cross-border capital flows, reducing dependence on
bank credit as a source of funding and, above all, lowering the risk of currency mismatches.
For foreign investors, it could actually be attractive to form diversified portfolios of
emerging market local currency debt issued by sovereign governments or developing country
corporations, with a return-to-risk that competes favourably with other major capital market
security indices.
88. Further development of these markets is desirable. First, developing countries bond
markets are still largely dominated by relatively short-term issues and, therefore tend to
correct currency mismatches but to increase maturity mismatches. Second, it has proved to
be much easier to develop large and deep local markets for public sector than for corporate
debt. As a result, large corporations continued to rely on external financing. To the extent
that such external financing is shorter term than that many developing country governments
are able to get in global debt markets, the overall debt structure of countries tends to
become shorter term–and therefore riskier. Indeed, the rollover of external corporate debt is
viewed as the major problem facing many emerging economies today. Third, many of these
markets are not very liquid. This problem has actually become more acute during the recent
market downswing. Fourth, although local bond issues did attract foreign investors, they
were largely or at least partly lured by the generalized expectations of exchange rate
appreciation that tended to prevail in many parts of the developing countries during the
recent boom. As the world financial crisis hit, there were large outflows of these funds, and
in this sense reliance on these short term portfolio flows did not correct but may have
enhanced pro-cyclicality of financing, much as short-term external bank debt did during
previous crises.
89. Therefore, although the development of local bond markets is a major advance in
developing country financing since the Asian crisis, its promise remains a partly unfulfilled in
terms of risk mitigation. It is important for developing country governments, with support
from international organizations to correct some of the problems that have been evident and
to continue investing in the development of deep and longer-term domestic bond markets.
Innovative Sources of Financing
90. For some time, the difficulty in meeting the UN official assistance target of 0.7 per cent
of GNI of industrial countries as official development assistance, as well as the need for
adequate funding for the provision of global and regional public goods (peace building,
fighting global health pandemics, combating climate change and sustaining the global
environment more generally) has generated proposals on how to guarantee a stable sources
of financing for these objectives.
91. This debate has led to a heterogeneous family of initiatives. A distinguishing feature of
developments in recent years is the fact that the old idea of innovative finance has started to
lead to action, with the launch in Paris, in 2006, of "the Leading Group on Solidarity Levies".
The Leading Group now involves close to 60 countries and major international
organizations.
110
92. Some of the initiatives that have been proposed encompass “solidarity levies” or, more
generally, taxation for global objectives. Some countries have already decreed solidarity levies
on airline tickets but there is a larger set of proposals. There have also been suggestions to
auction global natural resources–such as ocean fishing rights and pollution emission
permits–for global environmental programs.
93. The receipts from these innovative initiatives could be directed to support developing
countries to meet their development objectives, including their contribution to the supply of
global public goods, as well as international organizations that are active in guaranteeing the
provision of such goods. The existing taxes on airline tickets, for example, are being used to
finance international programs to combat malaria, tuberculosis and HIV-AIDS.
94. The suggestion of taxes that could be earmarked for global objectives has a long history.
To avert their being perceived as encroachments on participating countries’ fiscal
sovereignty, it has been agreed that these taxes should be nationally imposed, but
internationally coordinated. While universal participation is not indispensable, it would serve
the interest of development, as more resources would be raised. Some suggestions aim at
both raising funds for global objectives and mitigating a negative externality at the global.
Two suggestions deserve special attention: a carbon tax and a levy on financial transactions.
95. Since carbon dioxide is the main contributor to global warming, a tax on its emissions
can be defended on environmental efficiency grounds and would also have the advantage of
correcting a negative externality in addition to being a significant source of development
financing, which according to some estimates could reach as much as $130 billion per year.
However, carbon taxes should not be implemented in both developed and developing
countries. A uniform global carbon tax, even if introduced gradually, would mean that
developing countries will be taxed at several times the rate for industrial countries, as a
proportion of their respective GNPs. This would impose a disproportionate burden of
adjustment on developing countries, although per capita emissions of greenhouse gases in
developing countries are low compared with those in industrial countries. Carbon taxes will
also have adverse distributional impacts in developing countries. A high tax on an essential
good (e.g. energy, but also food or water) could render it unaffordable by lower income
groups. This would not only be regressive, but would also be socially unacceptable and
environmentally unpredictable.
96. An alternative to carbon taxes, which is now being used, is the auctioning emissions
rights. Emissions trading is one of the Kyoto Protocol mechanisms and has been
implemented by means of a European trading scheme which provides for an overall capping
of emissions. The mechanism makes pollution with greenhouse gases costly for the emitter
who has to acquire emission certificates. In this way, public revenues are generated which
can be used to finance both mitigation and adaptation to climate change in developing
countries, thus contributing to "climate justice".
97. Similar mechanisms can be designed to pay for environmental services. Such schemes
are operational locally in different areas of the world. They allow for consumers of a given
public good to compensate for some of the costs borne by those in charge of producing or
preserving it. For instance, downstream users of water can pay those who manage the
111
upstream forest to ensure a sustainable flow of this service into the future. It can be
envisaged that similar instruments could pay for the provision of global environmental
services, such as the preservation of rainforests.
98. Estimates of the revenues from a currency transaction tax range from $15 to $35 billion.
However attractive the tax might be in terms of revenue potential, its implementation is
constrained by a number of obstacles. Particularly, the tax base will have to be defined so as
to exclude certain transactions that provide very short-term liquidity to markets (e.g., when
this tax is applied at the national level, interbank lending is usually exempted) and special
treatment for derivatives to avoid double taxation. It will also have to be protected from
erosion, for even if all major financial centres participate, there is a risk that smaller centres
will attract an increasing volume of activity from those wishing to evade the tax. Finally,
strong opposition by a number of stakeholders would have to be overcome.
99. Alternatively, a levy on trade in shares, bonds and derivatives could be introduced.
Implementation would be easier than in the case of a currency transaction tax, as a small
number of participating countries suffices at the beginning. In later stages, over the counter
and currency trading could be included. Large stock exchange centres exhibit positive
agglomeration externalities; therefore a small tax would not lead to a flight of trade towards
alternative, smaller exchanges.
100. Another set of proposals rely on the use of financing mechanisms. One mechanism
that already has a long history of application is swaps of debt for development objectives. It
has been recently used in the Debt2Health initiative launched in Berlin in 2007, which
converts portions of old debt claims on developing countries into new domestic resources
for health. The International Finance Facility was proposed by the UK in 2003 to up-front
commitments for future flows of ODA, by issuing bonds backed by public or private sector
donors’ pledges. The first of these mechanisms, the International Finance Facility for
Immunization, is already in place.
101. Public-private sector partnerships can also be used to guarantee certain international
objectives. A mechanism of this type is the Advanced Market Commitments proposed by
Italy, through which government donors commit funds to guarantee the price of vaccines
once they have been developed, provided they meet a number of criteria on effectiveness,
cost and availability. This helps encourage pharmaceutical firms to focus on research into
neglected diseases which mainly affect poor countries.
102. Finally, from the outset, the Leading Group has been focusing on illegal financial flows
from developing countries, including those lost by developing countries through tax evasion
and other illegal means. It has set up for that purpose a task force on Global Financial
Integrity, under the direction of Norway. The importance of combating tax evasion has
already been underscored in chapter 3.
the need to develop a better system remains on the international policy agenda. The
international community needs to actively resume the effort to define the specific
mechanism to institutionalize the principles advanced here.
Foreign Debt Commission
76. The crisis also gives urgency to reform of institutional structures for debt relief, as an
increasing number of developing countries, especially the most vulnerable low-income
countries, may face difficulties in meeting their external debt commitments. This crisis
therefore gives urgency to these reforms. Unless these debts are managed better than they
have been in the past, the consequences for developing countries, and especially the poor in
these countries, can be serious.
77. Although, as argued above, there is a need for new procedures for restructuring
sovereign debt, this reform will take time, as it would require a new international treaty. In
the interim, something needs to be done to ensure that debts are better managed–and this
may be true even in the long run. It is important to take actions to manage debt better, so
that countries are not forced into default.
78. The United Nations should therefore set up a Foreign Debt Commission to consider
external debt problems of developing countries and economies in transition. The
commission, with balanced geographic representation and technical support from the
Bretton Woods, regional and other financial institutions, would address these issues and
provide advice on ways to enhance external debt crisis prevention and resolution.7 It would
also examine existing and advise on the design of better debt sustainability frameworks for
the international community to follow. It would help debt-distressed countries return to debt
sustainability, extend Paris Club-plus type approaches to new official creditors, set up an
interim mediation service, and craft on the basis of that experience more permanent debt
mediation and arbitration mechanisms.
Innovative Risk Management Instruments
79. The volatility of private capital flows to developing countries has generated increasing
demand for policies and instruments that would allow these countries to better manage and
minimize the risks associated with increasing international financial integration and, in
particular, to better distribute the risks associated with this integration among different
market agents. As has been demonstrated during past and current crises, the pro-cyclical and
7 See United Nations, “Doha Declaration on Financing for Development: outcome document of the Follow-up
International Conference on Financing for Development to Review the Implementation of the Monterrey
Consensus” (A/CONF.212/L.1/Rev.1), Doha, Qatar, 29 November -2 December 2008, paragraph 67.
107
herding behaviour of international capital flows tends to generate boom-bust cycles, which
are particularly damaging for developing countries, and it also reduces the scope they have to
undertake counter-cyclical macroeconomic policies. Moreover, many developing and
emerging countries borrow short term, in hard currencies, forcing them to bear the risk of
interest rate and exchange rate fluctuations. And finally, inadequate debt resolution
mechanisms impose high costs on developing countries.
80. In light of this, there have been a variety of ideas and proposals for introduction of
innovative financial instruments. The proposed instruments include tools that enable better
management of risks arising from the business cycle and fluctuations in commodity prices,
particularly GDP and commodity linked bonds and financial guarantees that have a countercyclical
element embedded in their structure. Promoting local currency bond markets has
also been seen as a way to enhance financial development and reduce the currency
mismatches that affect debt structures in developing countries.
81. GDP-linked bonds are conventional bonds that pay a low fixed coupon augmented by
an additional payment linked by a pre-determined formula to debtor country’s GDP growth.
This variable return structure links returns to the ability to service and thus reduces the
likelihood of costly and disruptive defaults and debt crises. The reduction of a country’s debt
service when the economy faces financing difficulties can also facilitate a more rapid
recovery, as it allows higher public spending in difficult times. For investors GDP-linked
bonds reduce the probability of default, and thus the costs of expensive renegotiation, and
offer a valuable diversification opportunity. Returns should be higher than with conventional
bonds.8
82. Since private financial markets are unlikely to develop these instruments autonomously,
because of the externalities associated with their introduction multilateral development banks
should take an active role in their development. In particular, these institutions could have
an active role as “market-makers”. The expertise developed by the World Bank as marketmaker
for the sale of carbon credits under the Kyoto protocol provides a precedent for these
activities. The World Bank and regional development banks could, for example, make loans
whose servicing would be linked to GDP. The loans could then be sold to financial markets,
either individually or grouped and securitized. Alternatively the World Bank or regional
banks could buy GDP-linked bonds that developing countries would issue via private
placements. The fact that major multilateral development banks became active in this type of
lending could, furthermore extend the benefits of adjusting debt service to growth variations
to countries that do not have access to the private bond market. GDP-indexed securities are
particularly appropriate for Islamic finance, as they can be made compatible with Sharia law
which prohibits charging of interest.
8 The introduction of these securities must overcome, however, some practical difficulties. One possible set of
concerns is associated with lags in the provision and frequent revisions of GDP data, as well as over the quality
of these estimates, but these issues should be easy to resolve through international standard setting and the
provision of technical assistance. More important in this regard is how to manage concerns that have been
raised about the liquidity of such instruments, especially when they are newly issued. Such concerns were
similarly raised when inflation indexed bonds were first introduced, but they are now accepted worldwide.
Governments and multilaterals might have to help create a thicker market.
108
83. There might also be alternative ways of ensuring flexible payment arrangements that
would allow automatic adjustment for borrowers during bad times. For instance, one
possibility is for coupon payments to remain fixed and the amortization schedule to be
adjusted instead. Countries would postpone part or all of their debt payments during
economic downturns; and they would then make up by pre-paying during economic
upswings. A historical precedent was set by the United Kingdom when it borrowed from the
United States in the 1940s. The Anglo American Financial Agreement was negotiated by
John Maynard Keynes and included a “bisque clause” that provided a waiver of 2% interest
payment in any year in which the United Kingdom foreign exchange income was not
sufficient to meet its pre war level of imports, adjusted to current prices.
84. Commodity-linked bonds can also play a useful role in reducing country vulnerabilities.
Examples of commodity-indexed bonds include oil-backed bonds, such as the Brady bonds
with oil warrants that were first issued on behalf of the government of Mexico. In such
instruments, the coupon or principal payments are linked to the price of a referenced
commodity. Again, it might be desirable for international institutions to help create a market
for such bonds.
85. However, the greater complexity of this instrument, in comparison with conventional
bonds, and the commodity price risk that the investor faces, may make commodity-linked
bonds expensive. Again, it might be desirable for international institutions to help create a
market for such bonds. While they are likely to be less useful than GDP-indexed bonds for
the growing number of developing countries that have a fairly diversified export structure
and therefore lack a natural commodity price to link to bond payments, they have the
decided advantage that the risk being “insured” through the bond is not one affected by the
actions of the country (i.e. moral hazard is less of a problem.)
86. Another way of addressing the problems created by the inherent tendency of private
flows to be pro-cyclical is for public institutions to issue guarantees that have countercyclical
elements. For example, Multilateral Development Banks (MDBs) and Export Credit
Agencies (ECAs) could introduce an explicit counter-cyclical element in all the risk
evaluations and the guarantees they issue for lending to developing countries. This requires
MDBs and ECAs to assess risk for issuing guarantees with a long-term perspective. When
banks or other lenders lowered their exposure to a country, MDBs or ECAs would increase
their level of guarantees, if they considered that the country’s long-term fundamentals were
basically sound. When matters were seen by private banks to improve, and their willingness
to lend increased, MDBs or ECAs could reduce their exposure. Alternatively, there could be
special stand-alone guarantee mechanisms for trade and/or long-term credit, for example
within multilateral or regional development banks, which had a strong explicit countercyclical
element. This could be activated in periods of sharp decline in capital flows and its
aim would be to try to catalyze private sector trade or long-term credits, especially for
infrastructure.
87. Finally, a number of developing countries have encouraged development of domestic
capital markets, and in particular local currency bond markets. These markets in fact
boomed after the Asian crisis, multiplying fivefold between 1997 and 2007 for the twenty
large and medium-sized emerging economies for which the Bank of International
Settlements provide regular information. This trend can be seen as a response of emerging
109
economies to the volatility and pro-cyclical bias of international capital flows and the
volatility of exchange rates. It can be viewed as a means of creating a more stable source of
local currency funding for both the public and private sectors, thereby mitigating the funding
difficulties created by sudden stops in cross-border capital flows, reducing dependence on
bank credit as a source of funding and, above all, lowering the risk of currency mismatches.
For foreign investors, it could actually be attractive to form diversified portfolios of
emerging market local currency debt issued by sovereign governments or developing country
corporations, with a return-to-risk that competes favourably with other major capital market
security indices.
88. Further development of these markets is desirable. First, developing countries bond
markets are still largely dominated by relatively short-term issues and, therefore tend to
correct currency mismatches but to increase maturity mismatches. Second, it has proved to
be much easier to develop large and deep local markets for public sector than for corporate
debt. As a result, large corporations continued to rely on external financing. To the extent
that such external financing is shorter term than that many developing country governments
are able to get in global debt markets, the overall debt structure of countries tends to
become shorter term–and therefore riskier. Indeed, the rollover of external corporate debt is
viewed as the major problem facing many emerging economies today. Third, many of these
markets are not very liquid. This problem has actually become more acute during the recent
market downswing. Fourth, although local bond issues did attract foreign investors, they
were largely or at least partly lured by the generalized expectations of exchange rate
appreciation that tended to prevail in many parts of the developing countries during the
recent boom. As the world financial crisis hit, there were large outflows of these funds, and
in this sense reliance on these short term portfolio flows did not correct but may have
enhanced pro-cyclicality of financing, much as short-term external bank debt did during
previous crises.
89. Therefore, although the development of local bond markets is a major advance in
developing country financing since the Asian crisis, its promise remains a partly unfulfilled in
terms of risk mitigation. It is important for developing country governments, with support
from international organizations to correct some of the problems that have been evident and
to continue investing in the development of deep and longer-term domestic bond markets.
Innovative Sources of Financing
90. For some time, the difficulty in meeting the UN official assistance target of 0.7 per cent
of GNI of industrial countries as official development assistance, as well as the need for
adequate funding for the provision of global and regional public goods (peace building,
fighting global health pandemics, combating climate change and sustaining the global
environment more generally) has generated proposals on how to guarantee a stable sources
of financing for these objectives.
91. This debate has led to a heterogeneous family of initiatives. A distinguishing feature of
developments in recent years is the fact that the old idea of innovative finance has started to
lead to action, with the launch in Paris, in 2006, of "the Leading Group on Solidarity Levies".
The Leading Group now involves close to 60 countries and major international
organizations.
110
92. Some of the initiatives that have been proposed encompass “solidarity levies” or, more
generally, taxation for global objectives. Some countries have already decreed solidarity levies
on airline tickets but there is a larger set of proposals. There have also been suggestions to
auction global natural resources–such as ocean fishing rights and pollution emission
permits–for global environmental programs.
93. The receipts from these innovative initiatives could be directed to support developing
countries to meet their development objectives, including their contribution to the supply of
global public goods, as well as international organizations that are active in guaranteeing the
provision of such goods. The existing taxes on airline tickets, for example, are being used to
finance international programs to combat malaria, tuberculosis and HIV-AIDS.
94. The suggestion of taxes that could be earmarked for global objectives has a long history.
To avert their being perceived as encroachments on participating countries’ fiscal
sovereignty, it has been agreed that these taxes should be nationally imposed, but
internationally coordinated. While universal participation is not indispensable, it would serve
the interest of development, as more resources would be raised. Some suggestions aim at
both raising funds for global objectives and mitigating a negative externality at the global.
Two suggestions deserve special attention: a carbon tax and a levy on financial transactions.
95. Since carbon dioxide is the main contributor to global warming, a tax on its emissions
can be defended on environmental efficiency grounds and would also have the advantage of
correcting a negative externality in addition to being a significant source of development
financing, which according to some estimates could reach as much as $130 billion per year.
However, carbon taxes should not be implemented in both developed and developing
countries. A uniform global carbon tax, even if introduced gradually, would mean that
developing countries will be taxed at several times the rate for industrial countries, as a
proportion of their respective GNPs. This would impose a disproportionate burden of
adjustment on developing countries, although per capita emissions of greenhouse gases in
developing countries are low compared with those in industrial countries. Carbon taxes will
also have adverse distributional impacts in developing countries. A high tax on an essential
good (e.g. energy, but also food or water) could render it unaffordable by lower income
groups. This would not only be regressive, but would also be socially unacceptable and
environmentally unpredictable.
96. An alternative to carbon taxes, which is now being used, is the auctioning emissions
rights. Emissions trading is one of the Kyoto Protocol mechanisms and has been
implemented by means of a European trading scheme which provides for an overall capping
of emissions. The mechanism makes pollution with greenhouse gases costly for the emitter
who has to acquire emission certificates. In this way, public revenues are generated which
can be used to finance both mitigation and adaptation to climate change in developing
countries, thus contributing to "climate justice".
97. Similar mechanisms can be designed to pay for environmental services. Such schemes
are operational locally in different areas of the world. They allow for consumers of a given
public good to compensate for some of the costs borne by those in charge of producing or
preserving it. For instance, downstream users of water can pay those who manage the
111
upstream forest to ensure a sustainable flow of this service into the future. It can be
envisaged that similar instruments could pay for the provision of global environmental
services, such as the preservation of rainforests.
98. Estimates of the revenues from a currency transaction tax range from $15 to $35 billion.
However attractive the tax might be in terms of revenue potential, its implementation is
constrained by a number of obstacles. Particularly, the tax base will have to be defined so as
to exclude certain transactions that provide very short-term liquidity to markets (e.g., when
this tax is applied at the national level, interbank lending is usually exempted) and special
treatment for derivatives to avoid double taxation. It will also have to be protected from
erosion, for even if all major financial centres participate, there is a risk that smaller centres
will attract an increasing volume of activity from those wishing to evade the tax. Finally,
strong opposition by a number of stakeholders would have to be overcome.
99. Alternatively, a levy on trade in shares, bonds and derivatives could be introduced.
Implementation would be easier than in the case of a currency transaction tax, as a small
number of participating countries suffices at the beginning. In later stages, over the counter
and currency trading could be included. Large stock exchange centres exhibit positive
agglomeration externalities; therefore a small tax would not lead to a flight of trade towards
alternative, smaller exchanges.
100. Another set of proposals rely on the use of financing mechanisms. One mechanism
that already has a long history of application is swaps of debt for development objectives. It
has been recently used in the Debt2Health initiative launched in Berlin in 2007, which
converts portions of old debt claims on developing countries into new domestic resources
for health. The International Finance Facility was proposed by the UK in 2003 to up-front
commitments for future flows of ODA, by issuing bonds backed by public or private sector
donors’ pledges. The first of these mechanisms, the International Finance Facility for
Immunization, is already in place.
101. Public-private sector partnerships can also be used to guarantee certain international
objectives. A mechanism of this type is the Advanced Market Commitments proposed by
Italy, through which government donors commit funds to guarantee the price of vaccines
once they have been developed, provided they meet a number of criteria on effectiveness,
cost and availability. This helps encourage pharmaceutical firms to focus on research into
neglected diseases which mainly affect poor countries.
102. Finally, from the outset, the Leading Group has been focusing on illegal financial flows
from developing countries, including those lost by developing countries through tax evasion
and other illegal means. It has set up for that purpose a task force on Global Financial
Integrity, under the direction of Norway. The importance of combating tax evasion has
already been underscored in chapter 3.