Supplementary memorandum submitted by
Mr John Roberts, Head of CAPE, ODI
PROPOSAL FOR AN INTERNATIONAL DEVELOPMENT
1. The main points made in this note are
1.1. If there is a large increase in development
assistance as proposed by the Zedillo Panel it is desirable that
there should be an impartial and respected body, whose legitimacy
is not questioned, to determine the uses of the extra aid and
how it should be administered. A suitably constituted International
Development Trust Fund might fulfil this role.
1.2. The quickest way to mobilise that part
of the proposed $50 billion per annum of additional development
assistance that will not come in the form of conventional ODA
(which donors pledged to increase on the eve of Monterrey) would
be guaranteed market borrowing, as proposed by the Chancellor.
1.3. Guaranteed market borrowing in amounts
that seem likely to be required (some $40-45 billion per annum
in current prices post 2005) would soon give rise to an embarrassingly
large burden of debt for guarantor countries that might impair
their ability to sustain and increase their conventional ODA effort.
1.4. It is therefore highly desirable that
other non-conventional sources of development assistance, such
as SDRs or hypothecated international taxes, be agreed and implemented
as soon as possible so that reliance on market borrowing is reduced
and then eliminated, if at all possible before 2010.
2. In his speech to the New York Federal
Reserve Bank on 16 November 2001 the Chancellor of the Exchequer
outlined ways in which international donors could meet the additional
$50 billion per annum target for additional official development
assistance that the Zedillo Panel thought necessary to meet the
Millennium Development Goals.
3. The Chancellor proposed that:
The richest countries should commit
themselves to a substantial increase in their ODA up to and beyond
the year 2015;
In order to meet the $50 billion
per annum target contributions additional to this conventional
ODA should be sought through market borrowing with the guarantees
of donor governments or against other forms of collateral;
The $50 billion should be disbursed
in a balanced way through existing bilateral, multilateral and
civil society channels, and used to support developing countries'
poverty reduction strategies;
One way of handling the additional
funds would be to create an International Development Trust Fund
overseen by the World Bank, the IMF and perhaps some individual
4. No further details of how these proposals
would work in practice have been made public. However, it is possible
to construct outline scenarios describing their operation, both
as regards the uses to be made of the additional aid flows and
as concerns their financing. This note offers a rationale for
an International Development Trust Fund and an illustration of
the implications of financing the Fund by market borrowing.
5. The volume of net ODA stagnated in the
1990s. Its average level in 1989-1990 was $57 billion and in 1999-2000
it was $56 billion.
Increasing this level of flow by $50 billion (90 percent) would
place strain on existing management capabilities in both donor
agencies and in aid recipient countries. It would very likely
imply some alteration in current priorities in developing countries'
poverty reduction strategies and their aid use. It would require
that close attention be paid to increasing recipients' expenditure
programme management capacities. Recipients would also have to
take care that the macroeconomic effects of higher aid did not
damage their longer term growth prospects.
6. The prospect of very much larger aid
flows would create particular pressures on multilateral aid institutionsto
which bilateral donors would probably look to handle a significant
share of the additional flows. There would be rivalry between
institutions over the shares of the extra aid which they should
control. Those institutions such as those in the UN group which
now regard themselves as underfunded would expect large increases
in their endowments. On the other hand, the World Bank and the
regional development banks which are used to disbursing large
volumes of concessional finance to developing countries would
be looked to by some donors as the natural administrators of the
bulk of the additional resources.
7. Special purpose trust funds set up to
meet particular development challenges and to provide larger volumes
of "global public goods" would probably emerge and increase
in number to take advantage of the prospect of an increase in
aid volume. An example of these is the recently created Global
Fund for Malaria, TB and HIV/AIDS.
8. A quantum increase in aid would therefore
make necessary the creation of some means of adjudicating between
competing claims by institutions to manage the resources and to
determine thematic and geographical priorities for their use.
One purpose of an International Development Trust Fund would be
to stand as impartial arbiter in this process. The Trust Fund
would have primary responsibility for allocating to multilateral
and other international channels those additional resources that
bilateral donors do not themselves wish to administer. To do this
it would have to form a collective view on: (a) priorities in
using aid for poverty reduction, (b) assessing aid effectiveness
and developing countries' absorptive capacity, (c) the most appropriate
forms of support, and (d) which institutional channels are most
apt to administer this support. The Trust Fund would also have
to take a view on the shares of resources to be channelled to
global programmes and to national development programmes.
9. In the spirit of Monterrey a wide spectrum
of countries would have to be represented in the council of the
Trust Fund. It would create unfortunate North-South and inter-institutional
tensions if the Trust Fund were felt to be the creature only of
the major donors or only of particular multilateral agencies.
Suggestions have been made in the past that the product of novel
sources of development finance should be administered by the UN.
However, the UN is itself an interested party and there is at
present no UN body able to take a sufficiently impartial and detached
view of needs and capabilities.
10. One of the main proposals in the Chancellor's
speeches of late 2001 was in fact translated into specific pledges
in the approach to the Financing for Development conference in
Monterrey. This is that the donors should commit themselves to
substantial increases in their aidprior to and beyond the
year 2015. The European Community and its member states committed
themselves to increase their aid disbursements by $7 billion per
annum over the 2004-06 period; and the US promised to increase
its aid effort by $5 billion per annum in the course of the same
period. Other significant donors such as Canada and Japan also
indicated their willingness to increase their aid flows.
11. If these pledges are fulfilled they
should in due course give rise to additional net flows of assistance
of some $12-14 billion per annum at today's prices. Furthermore,
the pledges before Monterrey were couched in terms of increases
in the percentages of donors' Gross National Income devoted to
aid. This should mean that the higher levels of official flow
envisaged will be made to grow at least as fast in nominal terms
as nominal GNIsay 5 per cent per annum.
12. These welcome pledges will however still
leave a large volume of finance to be raised by other means if
the $50 billion target is to be attained, viz $36-$38 billion
at today's prices. To raise these resources innovative sources
of financing, lying outside aid donors' conventional budgets,
will have to be sought.
13. The Chancellor's proposal for mobilising
extra-budgetary resources is that there should be market borrowing
against guarantees (or other forms of collateral) provided by
donor governments. Other suggestions include the emission of new
SDRs and the raising of various forms of international tax, eg
the Tobin Tax on international transactions, a carbon tax and
an air transport tax. It should be stressed that these suggestions
are not mutually exclusive. In fact there are practical and presentational
advantages in combining two or more of them.
14. The modalities of the market borrowing
envisaged by the Chancellor have not been spelled out. However,
it is not difficult to construct a financial scenario to illustrate
the magnitudes involved. For this purpose it is convenient to
assume that the International Development Trust Fund will have
a legal status and intergovernmental guarantees and, as such,
will have the ability and responsibility of issuing bonds in its
own name up to the amount required to fill the shortfall in development
finance below the $50 billion target.
15. Other assumptions used in the scenario
are that: (a) the interest yield on IDTF bonds will continue indefinitely
to be a nominal 5 per cent (roughly the current yield on the most
creditworthy US and European long term bonds), (b) the IDTF will
be able to roll over its debt for as long as it wishes, (c) the
IDTF will pay no interest,
but will roll interest due up into new debt, the IDTF will provide
development finance in the form of credits on highly concessional
(d) inflation will remain of the order of 2-2.5 per cent.
16. The table in the Annex projects (Col
1) additional aid disbursements reaching the inflation-adjusted
equivalent of $50 billion in 2005. The conventional ODA share
of this (Col 2) rises more slowly but, by 2007, contributes $15
billion to the total.
This increase raises the ratio of OECD donors' ODA to GNI modestly
from 0.22 per cent in the recent past up to 0.27 per cent.
Overall additional disbursements (Col 1) rise at the rate of inflation,
while additional conventional ODA rises, post 2007, at the rate
of GNI growth (assumed to be 5 per cent per annum). The shortfall
between the inflation-adjusted additional disbursement target
and additional conventional ODA is borrowed in the markets. The
net market borrowing requirement (of approximately $40-45 billion
per annum in nominal terms) is shown in Col 4.
17. The projection shows that, if interest
on IDTF bonds is rolled up into new debt and no debt is repaid,
the debt of the Trust Fund mounts rapidly, reaching $400 billion
by 2010 and $750 billion in nominal terms by 2015 (Col 6). Debt
service payments due to the IDTF from developing country borrowers
(Col 3) are puny compared with annual interest payments on accumulated
ITDF debt (Col 5), and have insignificant impact on the size of
18. If it were decided to call a halt to
market borrowing in 2010 and to repay the IDTF's accumulated debt
in equal annuity instalments over the following 15 years the annual
cost to guarantor governments would be $38.5 billion (which could
be partly offset by interest and amortisation payments from aid
recipient countries rising from $3 billion to $25 billion over
the periodCol 4). If the IDTF were to continue borrowing
up to and including 2015 guarantor governments would have to pay
$72.25 billion per annum in order to expunge the debt by 2030
(partly offset by debt service payments by recipients rising from
$10 billion to $28 billion over the period).
19. It goes without saying that the debt
and its servicing would be less onerous if conventional ODA flows
increase above the levels shown in Col 2 of the Annex table, thus
limiting the need for market borrowing. Koos Richelle, Director-General
of DGDEV, has proposed a mechanism for raising conventional ODA
flows from EU member states. This consists of asking those member
states whose aid disbursements are less than 0.7 per cent of GNI
to raise them at least to the current EU average (0.33 per cent)
by 2006. If the mechanism is enforced from now on this ratio will
increase to 0.39 per cent in 2006 (and potentially rise thereafter
towards 0.7 per cent of GNI).
20. The following table illustrates the
effect on the need for market borrowing and on IDTF debt of a
staged increase of the ratio of all OECD donors conventional ODA
to GNI from its current level of 0.22 per cent to a ceiling of
0.4 per cent, reached in 2007.
The results are compared with more modest increases announced
before Monterrey which take the ODA/GNI ratio only up to 0.27
per cent. The higher level of conventional ODA would reduce IDTF
debt very significantly by nearly 30 per cent in 2010 and by 50
per cent in 2015, compared with the lower level.
21. An ODA/GNI ratio of 0.4 percent, while
modest and attainable for the EU, would be extremely challenging
for the US (whose ODA/GNI ratio would almost have to quadruple),
and also challenging for Japan (whose ratio would have to rise
by nearly 70 per cent from its reduced 2001 level). On present
form, therefore, it is regrettably improbable that conventional
aid will rise to this level in the present decade, even if the
EU effectively implements the Richelle mechanism.
| 2007 projectedprojected
|| 2010 projectedprojected
|| 2015 projectedprojected
|ODA/GNI ratio (%)||0.22
Total net conventional ODA ($billion at current prices)
IDTF debt ($ billion at current prices, end of period
22. Although these scenarios are only illustrative they
strongly suggest that, without an early and improbably large increase
in conventional ODA, ITDF debt could quickly reach a magnitude
at which debt service could have a negative effect on donors'
willingness to provide conventional ODA in the longer term. Guarantor
governments might find themselves unable both to meet their obligations
to pay down ITDF debt and to sustain even their modest Monterrey
pledges to increase ratios of ODA to GNI. This could bode ill
for development assistance volumes in the years after 2015.
23. For this reason it would be advisable for the ITDF
to seek to mobilise sources of non-conventional additional financing,
other than market borrowing, as soon as possible. It will not
be easy to reach international agreement on creating new SDRs
or raising new hypothecated international taxes of the kind mentioned
in paragraph 12 above. Many more parties would be required to
give their assent than would be needed to start guaranteed market
borrowing. Entrenched institutional resistance might well be encountered.
The yield on some possible forms of international tax would probably
be limited. However,
in the interest of sustaining the higher levels of development
assistance recommended by the Zedillo Panel it is important that
one or more innovative sources of finance be tapped at the very
latest by 2010, thus preventing a build-up of ITDF debt which
could, within a few years, have seriously negative consequences
for net assistance flows to poor countries.
Mr John Roberts
Head of Centre for Aid and Public Expenditure (CAPE), ODI
At 1999 prices and exchange rates-cf "Development Cooperation-2001
Report" Statistical Annex Table 9, OECD Development Assistance
Committee, 2002 Back
Eg by counteracting possible inflationary and exchange rate overvaluation
George Soros in "on Globalisation" (2002) proposes that
additional development finance should be allocated to a range
of such special purpose funds. Back
UNDP could be an arbitrator if it returned from its present role
as technical cooperation provider and in-country UN agency coordinator
to its original role as adjudicator within the multilateral system. Back
At least within the period of the projection, because interest
on IDTF debt is likely to compete for resources with transfers
to developing countries. Back
Comprising a 10 year grace period, a 40 year repayment period
and an interest rate of 0.75 per cent. Back
Ie, the $12 billion per annum pledged by the EU and the US prior
to Monterrey, increased by the rate of inflation and augmented
by additional conventional aid from other OECD donors. Back
OECD (DAC) projects an ODA/GNI ratio of only 0.24 per cent in
2006 if pre-Monterrey pledges are fully implemented, based perhaps
on a low estimate of what EU actions will yield or on a high assumed
rate of erosion of the pledges by inflation. Back
The 2001 unweighted average ODA/GNI ratio of OECD donors is 0.4
per cent. Back
The ODI Memorandum to the International Development Committee
on "Financing for Development", estimated, for example,
that an international air transport tax might only yield $6-7
billion per annum. Back