World Bank: Funding IDA's Assistance Program

Report for Congress
World Bank: Funding IDA’s Assistance Program
Updated May 21, 2002
Jonathan E. Sanford
Specialist in International Political Economy
Foreign Affairs, Defense and Trade Division

Congressional Research Service ˜ The Library of Congress

World Bank: Funding IDA’s Assistance Program
During the past decade, the World Bank’s concessional loan facility, the
International Development Association (IDA), has relied increasingly on loan
repayments (“reflows”) to fund its assistance program. By committing future reflows
to fund current projects as they are disbursed, the World Bank and its member
countries have raised to nearly 40% the share of IDA’s current loan program funded
with reflows. In future years as the stock of future anticipated reflows becomes
increasingly obligated to pay for projects approved earlier, IDA will be relying more
on its reserves to fund its future lending program. Under current policies, IDA’s
cash balance will decline from $11.78 billion (2001) to about $1.4 billion (2013) and
then will roughly stabilize for several decades. The World Bank and its member
countries anticipate that loan repayments will account for 70% of the resources
needed to fund IDA’s loan program between 2032 and 2041.
This effort is complicated by the parallel effort of the World Bank and its
member countries to provide debt forgiveness to heavily indebted poor countries
(HIPCs). To date, the World Bank has forgiven over $9 billion in IDA debt owed by
HIPC countries. It has booked most of this debt but – because most of the
repayments are not due for many years – it has not yet realized most of the loss. The
World Bank’s plan to make IDA mostly self-sustaining in the future assumes that
donor countries will reimburse its HIPC debt forgiveness in full. If the forgiven loan
reflows are not restored, IDA will deplete its cash balance in 2009 and need to start
shrinking its aid program. The cost of funding the lost reflows will be roughly $500
million to $600 million a year for the next 20 years. The United States typically has
provided about 20% of the money to fund HIPC multilateral debt relief, thus making
this an important issue for Congress.
Likewise, the Administration and others have proposed that IDA adopt a major
grant program for poor countries. The scope and details of the plan are currently
under negotiation by donor countries. A grant program will need to be funded by
contributions from donor countries or IDA will need to start reducing the size of its
future program in anticipation of lost reflow income. Estimates of the size of the cost
of the Administration’s proposed grant program differ. Much of the difference in the
estimates is due to the methods used for presenting the estimates rather than to major
differences in the size of the expected costs. CRS and GAO both estimate that the
cost of the Administration’s grant program could be funded in full if donors
contributed between 1.6% and 1.8% more each year (compound interest). As such,
they would need to contribute between $64 billion and $71 billion more during the
next 40 years. This would be a 40% to 45% increase in the donors’ overall
contributions to fund the IDA program. Based on prior practice, the United States
would likely be asked to provide 20% of that total.
Basically, the goal of making IDA more self-funding through reflows and the
goal of reducing recipient countries’ repayment obligations are incompatible. The
United States and other donor countries may need to decide which goal to favor –
whether to fund IDA debt forgiveness and grants with increased contributions or to
allow the overall size of IDA’s assistance program diminish in future years.

In troduction ......................................................1
Funding IDA’s Current Program......................................1
Funding IDA’s Program: Grants......................................7
Balancing Multiple Goals..........................................10
List of Tables
Table 1. Projected IDA Cash Flow and Cash Balance, 2002-41..............3
Table 2. Projected IDA Cash Balance, 2002-2021
Without HIPC Reimbursement...................................5
Table 3. Projected IDA Cash Balance, 2002-2041
With Unfunded Grants.........................................8

World Bank: Funding IDA’s Assistance
The International Development Association (IDA) is the component of the
World Bank that makes concessional loans to the world’s poorest countries. IDA
gets most of the new money it uses to fund its assistance programs from contributions
by donor countries such as the United States, European countries, and Japan. IDA
lends on easy repayment terms, with no interest charge and repayments stretched over
35 to 40 years, including a ten year grace period. It levies a 3/4 of 1% service charge
on the disbursed balance for its loans to help offset IDA administrative expenses.
This paper analyzes the way IDA’s assistance program is funded and at the
effort by donor countries and the World Bank to have loan repayments (“reflows”)
carry an increased share of IDA costs. It also explores the potential long-term effects
that debt forgiveness and a major grant program might have on IDA’s finances. IDA
has forgiven substantial amounts of debt owed to it by heavily indebted poor
countries (HIPCs). The question whether the costs of that debt relief will be borne
by IDA or by donor countries has not yet been resolved. Likewise, President Bush
has proposed that IDA institute a major program of grants to poor countries.
Currently, IDA makes only loans. The costs of this initiative and the way it would be
funded have likewise not yet been determined. In its examination of the two issues
– debt forgiveness and grants – the paper discusses the likely cost the United States
and other donors would need to bear if they chose to support the costs of IDA debt
forgiveness or IDA grants themselves.
For fiscal 2003, the Administration has requested appropriations of $874 million
for contribution to IDA. Contribution levels will likely remain at this level, or
perhaps increase, in future years. In recent years, Congress has also appropriated
about $775 million to fund multilateral debt forgiveness to HIPC countries.
Proposals for further HIPC debt relief and a new IDA grant program are currently
under consideration and will likely require additional U.S. contributions in future
years. Future U.S. contributions or changes in IDA or the HIPC program will need
the consent of Congress. Shifts in the way these programs are funded could have a
substantial impact on prospective levels of support.
Funding IDA’s Current Program
Every three years, as IDA’s stock of uncommitted resources declines, donors
meet to negotiate terms for a replenishment of IDA’s resources. Since IDA was
created in 1960, there have been twelve such replenishments. A new replenishment
plan (IDA 13) -- to fund IDA operations during the next three year period – is

currently being negotiated. Legislation to authorize U.S. participation and U.S.
contributions to the plan likely will be submitted to Congress in June 2002.
In the past, most of the money IDA had available to lend during each
replenishment period (the three years funded by a replenishment plan) typically came
from donors. Funds from loan repayments were also used to support new loans, but
the amount received annually from this source was relatively small. During the past
decade, however, the volume of reflows has increased substantially. In 1992, the
IDA donor countries adopted a new plan (in IDA 10) to enhance through “advanced
commitment” the role that reflows will play in IDA’s operations. IDA previously
waited until reflows were in hand before adding them to the stock of resources
available to finance future loans. Under the advanced commitment plan, reflows
received during the next decade are counted as support for an IDA replenishment
plan if they are used to pay for disbursements on projects approved during that
replenishment period. Because IDA only releases funds as implementation goes
forward, the disbursements for an IDA project may be stretched out over a six to
eight year period.
Even though the amount being received as reflows during a three-year
replenishment period is relatively small, the share of IDA disbursements funded by
reflows has increased substantially. In IDA 10, IDA loan commitments made during
the period 1994 to 1996 were funded with $16.3 billion in new contributions from
donors and $3.1 billion in advanced commitments of reflows.1 In other words,
reflows accounted for about 16% of the resources used to fund IDA 10. In IDA 11
(1997 to 1999), IDA’s loan program received $9.86 billion in new contributions
from donors and $6.5 billion in advanced commitment of reflows. (Another $1.63
billion in IDA 11 projects was funded with money carried over from earlier
replenishments.) On this basis, omitting the carryover to avoid double counting,
reflows accounted for 40% of IDA disbursements during the IDA 11 period.
However, since actual reflows received during those three years amounted to only
$2.39 billion, IDA needed to use $3.1 billion in future reflows to fund disbursements
on IDA 11 projects.
In IDA 12 (2000 to 2002), new contributions amounted to $12.44 billion while
reflows covered $6.4 billion in future IDA 12 disbursements. In effect, reflows
provided about 34% of the funds needed to pay for IDA 12 loans. As IDA was
scheduled to receive $4.46 billion in reflows during that three year period, IDA
would need $1.94 billion in future reflows to pay IDA 12 costs. In IDA 13, currently
under negotiation, donors (including the IBRD) would provide $13.7 billion in new
money and “internal resources” (reflows plus reserves, a point to be discussed later)
would comprise $8.9 billion. On this basis, reflows and other “internal” IDA
resources would account for over 47% of IDA 12's total cost.

1Throughout this paper, one SDR is valued at $1.253. This makes easier the comparison of
numbers from one replenishment to the other. The World Bank has used this procedure in
its consultations with member countries about long-term IDA funding plans.

IDA’s ability to commit future reflows to fund its current loan program is
facilitated by the nature of IDA reflows. On average, 95% of the amount due from
borrowers is repaid on time and in full. Most of the countries not paying have (in
effect) no national government or their government is seriously estranged from the
world system.
Table 1. Projected IDA Cash Flow and Cash Balance, 2002-41
(Billions of US Dollars)
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Commitment s 6.28 7.35 7.35 7.35 7.87 7.87 7.87 8.43 8.43 8.43
Cont r ibut ions -- 4. 51 4. 51 4. 51 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39
Contrib old4.002.001.621.62------------
Net Op Income0.810.830.860.890.890.820.760.750.720.70
Ref lows 1.30 1.40 1.60 1.80 1.90 2.10 2.30 2.50 2.70 2.90
Total income6.559.
Disbursmts new1.432.914.305.486.407.157.708.108.21
Disbursmnts old6.497.516.085.324.
Total Costs6.939.399.4510.0910.079.908.769.559.438.24
Net annual cash-0.38-0.20-0.40-0.80-2.40-2.10-0.81-1.39-1.100.29
Balance 11.4011.2010.8010.007.605.504.703.302.202.50
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Com m it m ent s 9 . 0 2 9 . 0 2 9 . 0 2 9 . 6 6 9 . 6 6 9 . 6 6 10. 34 10. 34 10. 34 11. 07
Cont r ibut ions 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39
Net Op Income0.740.730.740.800.850.870.940.960.991.07
Ref lows 3.30 3.50 3.80 4.00 4.30 4.50 4.80 5.10 5.30 5.60
Total income8.989.199.519.7810.0910.3410.7211.0511.2911.69
Disburs new8.408.588.748.989.199.369.619.8310.0210.29
isb correction0.730.45-0.510.810.26-0.600.730.52-0.540.67
Total Costs9.689.598.8010.3810.009.3310.9210.9510.0911.58
Net annual cash-0.70-0.400.71-0.600.091.01-
Balance 1.801.402.101.501.602.602.402.503.703.80
2022 2023 2024 2025 2026 2027 2028 2029 2030 2031
Com m it m ent s 1 1. 07 11. 07 11. 85 11. 85 11. 85 12. 69 12. 69 12. 69 13. 58 13. 58
Cont r ibut ions 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39
Net Op Income1.
Ref lows 4.83 5.19 5.51 5.87 6.22 5.43 5.79 6.16 6.52 6.88
Total income10.9611.3511.7112.1112.5211.7812.1512.5412.9213.30
Disburs new10.5310.7211.0111.2711.4811.7912.0712.2912.6212.92
Disb corection--------------------
Total Costs11.1711.3711.6811.9512.1812.5112.8013.0413.3913.70
Net annual cash-0.21-
Balance 3.603.583.613.764.103.382.732.231.771.37
2032 2033 2034 2035 2036 2037 2038 2039 2040 2041
Com m it m ent s 1 3. 58 14. 54 14. 54 14. 54 15. 57 15. 57 15. 57 16. 67 16. 67 16. 67
Cont r ibut ions 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39 4. 39
Net Op Income1.
Ref lows 6.00 6.39 6.78 7.20 7.62 7.34 7.79 8.24 8.68 9.14
Total income12.4412.8213.2513.7114.1713.9414.4314.9215.4115.92
Disburs new13.1613.5213.8314.0914.4714.8115.0815.4915.8516.15
Disb correction-0.15-1.52-1.42-1.24-1.17-1.78-1.57-1.50-1.40-1.22
Total Costs13.8112.8213.2513.7114.1813.9314.4314.9315.4115.91
Net annual cash-1.370.00-0.000.00-0.010.01-0.00-0.00-0.000.00
Balance - 0. 00 - 0 . 0 0 - 0. 00 0. 00 - 0 . 0 0 0 . 0 0 0 . 0 0 - 0. 00 - 0 . 0 0 - 0. 0

Source: Created by CRS using data supplied by World Bank, as discussed in text.
At the beginning of July 2001, IDA had uncommitted financial resources on
hand totaling $11.78 billion. It also had $9.24 billion in donor pledges in non-
negotiable non-interest bearing notes which had not yet been encashed. IDA cannot
require that its donors turn their contribution pledges into actual money right away.
Rather, the donors’ notes must be drawn down over a 9 year period according to a
specified schedule. IDA’s cash balance (uncommited financial resources) has grown
gradually over the years because contributions have come in at the scheduled rate,
reflows have increased, and disbursements seem to have been somewhat less than
originally expected. The latter appears to be due partly to difficulties the Bank has
had implementing complex projects in difficult contexts (IDA had over $20 billion
in undisbursed loans commitments in 2002, some 19% of all loan commitments to
date) and partly to the dip in IDA loan commitments which occurred in the late


Table 1 seeks to present a picture of IDA’s financial situation for the next 40
years. As with all projections, the data are strongly dependant on the accuracy of the
original numbers and the strength of the underlying assumptions. The calculations
here are based on data or assumptions provided by the World Bank.2 Contributions
include planned future transfers from IBRD net income as well as contributions from
governments. Operations and administrative costs would increase by 2.3% annually,
per the expected rate of inflation. Contributions from donors would remain the same
in nominal terms. Investment held in the cash balance are assumed to earn 5%
annually. Income from the IDA service charge would continue to be 3/4 of 1% of the
disbursed balance. Disbursement figures were the only data not specifically defined
by World Bank data or assumptions. The U.S. General Accounting Office (GAO)
provided estimates for annual disbursements from future replenishments, based on
standard World Bank procedures. The author adjusted disbursement levels for years
2002 to 2021 (funded with new and old money) in order to keep IDA’s cash balance
each year at the level projected by the Bank. Future disbursement data reflows ( for
2022 to 2041) were generated by the assumptions built into the model. Estimates for
future reflows were calculated using the disbursement data for the first two decades
and standard disbursement and repayment formulas.
According to the World Bank, IDA will fund much of its future loan program
through substantial reductions in its cash balance. Over the next 12 years, IDA plans
to reduce its cash balance from $11.78 billion (2001) to $1.4 billion (2013). During
those years, IDA will disburse $1 billion or more annually than it receives in income.

2 World Bank. IDA Funding Scheme and Financial Projections: IDA13 Replenishment.
Powerpoint slides distributed to IDA donor countries, late fall 2001. Unpublished.

IDA’s cash balance will remain favorable through most of the 2020s, but will begin
sinking in the last years of the decade. According to the model, disbursements will
need to be constrained in the 2030s if IDA is not to run out of money. IDA’s cash
balance is its protection against contingencies, its “margin of error” against
unexpected events. Whether IDA’s future assistance program will be funded
substantially by “reflows” in the next two decades is a matter of semantics, given the
many sources of funds that have contributed to the current cash balance. A major
share of IDA’s reflows during the next decade have already been committed to fund
earlier replenishments.
The World Bank assumes that IDA resources in real terms will remain the same
size they are now for the next 40 years. Loan disbursements and administrative costs
will grow by 2.3% annually (per expected inflation), while donor contributions would
remain flat in nominal terms. The Bank presumes that, all else being equal, loan
disbursements and other non-donor sources of revenue will provide 70% of the
money IDA will need to fund its loan program during the last decade (2032 to 2041)
of this period.
Funding IDA’s Program: HIPC
This is the point in the present discussion where the issue of debt forgiveness
to poor countries through the HIPC program and possible IDA grants becomes
important. Both initiatives would reduce the amount of reflows that IDA receives in
future years – shifting the burden either to donors (a need for increased
contributions) or to recipients (a smaller IDA assistance program). The data provided
by the World Bank (shown in Table 1) assume that there will be no significant IDA

grant program and that IDA is reimbursed in full (by donor countries) for all the
debts it has forgiven the HIPC countries.3
Through December 31, 2001, the World Bank has forgiven or is forgiving
$10.2 billion in debt owed to IDA through the HIPC program. To offset some of
these costs, it received $778 million in reimbursements from the HIPC trust fund.
The cost to IDA for debt forgiveness occurs not in the year when the forgiveness is
announced, but in the year when IDA would have otherwise received the loan
repayment. IDA has booked a net loss of $8.6 billion in HIPC forgiveness (more will
be booked in the future), but to date it has only realized a small portion of the loss.
Table 2. Projected IDA Cash Balance, 2002-2021
Without HIPC Reimbursement
(Billions of U.S. Dollars)
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
HIPC lost reflows0.320.360.440.480.520.550.580.580.550.57
Balance 11. 29 10. 82 9. 97 8. 65 5. 33 2. 24 0. 39 - 1 . 9 6 - 3. 90 - 4 . 3 7
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
HIPC lost reflows0.580.590.570.550.550.550.550.550.450.35
Balance - 5. 83 - 6 . 9 6 - 6. 94 - 8 . 2 5 - 8. 83 - 8 . 5 0 - 9. 43 - 10. 05 - 9 . 4 7 - 9. 96
Source: Created by CRS using HIPC reflow data from World Bank and Table 1 data.
Table 2 shows (based on Table 1) the effect of that HIPC debt forgiveness will
have on IDA’s cash balance if the lost reflows from HIPC are not reimbursed and
IDA has to absorb the cost from its own resources. For simplicity, only the results of
the calculation are presented.4 IDA’s cash balance will become negative in 2009 in
that situation. The difference between the annual balances in Table 2 and those in
Table 1 represent the cumulative amount IDA would need to receive from some other
source if it were to hold the size of its aid program unchanged despite the shortfall
in reflows. The World Bank was unable to provide data on the impact that HIPC
would have on IDA reflows beyond 2021. The World Bank says that it plans to
contribute between $200 million and $220 million to HIPC in the next four years

3 World Bank. IDA Funding Scheme..., p. 9. Among the specific assumptions for the
Bank’s funding plan is a statement that “Costs of HIPC debt service forgiveness [are]
covered by IBRD net income pledge and subsequently by additional donor contributions.”
Future IBRD contributions to IDA are included in the model but not future IBRD
contributions to HIPC. The Bank assumes that coverage of past and future HIPC
forgiveness by the IBRD and by new donor contributions is a prerequisite (not included in
the data) to the arrangements specified in its new 40 year plan.
4 Table 2 shows the key changes that would occur in Table 1 if the reduction in reflows
caused by HIPC debt forgiveness were added to IDA costs. If that additional cost were
added to Table 1, the cash balance (bottom line) figure would change to reflect the data
shown in Table 2.

(through 2005), completing its pledge to provide substantial IBRD contributions to
that program.5 This amount is included in the income calculation for Table 2.
There appear to be three ways IDA can deal with that situation. First, it can
shrink the size of its annual IDA aid program by an amount equivalent to the lost
HIPC reflows. Second, the lost reflows can be reimbursed to IDA by donor
countries. Third, IDA can seek funds from some other source, in the World Bank or
in another international financial institution to cover the shortfall.
Reducing the size of the IDA program could save money but it could also have
negative effects on recent HIPC graduates and other poor countries. If donors wish
to offset the lost HIPC reflows, they would need to contribute $8.64 billion over 20
years.6 If they fund the loss each year as it occurs, they would need to contribute the
amounts shown in Table 2. Alternatively, if they chose to fund the lost revenues up
front, they could make a one-time payment of $5.6 billion. Invested at 6.3%, this
would generate enough revenue (principal and interest) to offset the lost revenues in
each future year. Based on recent precedents, the likely cost to the United States in
either case would be about 20% of the total.
The World Bank has substantial assets in its reserves. However, these are assets
of its near-market loan facility, the International Bank for Reconstruction and
Development (IBRD). Transferring them to IDA may be difficult from a legal and
political perspective. It would require a super-majority vote of the membership, it
may engender lawsuits by Bank bondholders, it would likely drive up interest rates
the Bank would need to pay to borrow funds, and – if a distribution of assets is
triggered – it would require an appropriation by Congress to transfer the U.S. share
of those assets (several billion dollars) to the IDA.
Funding IDA’s Program: Grants

5 The World Bank promised its member countries that it would contribute $2.15 billion in
net present value (NPV) terms over time to the HIPC trust fund. To date, the World Bank
has transferred $1.4 billion from IBRD net income (not reserves) to the HIPC trust fund. The
remainder would be contributed by 2005. See: World Bank. Heavily Indebted Poor
Countries (HIPC) Initiative: Status of Implementation. March 22, 2002. IDA/SecM2002-
0155. Prepared for the Development Committee Meeting, April 21, 2002. See paragraph
23. Also available from the Bank web site. The Bank says that this will permit the IDA
component of the HIPC Trust Fund to reimburse IDA for IDA debt through the end of IDA
13. The Bank says that it expects that IDA donors will consider the future needs for funding
IDA’s HIPC costs in the IDA 14 negotiations. During IDA 13 (upcoming), the IDA donors
also reportedly plan to contribute $627 million to cover such costs during the IDA 13
replenishment period. This amount is not included in the income data for Table 2. The
donor countries reportedly plan to cover subsequent shortfalls in HIPC funding on
a pay-as-you-go basis, though final agreement on this point has been postponed until
the negotiations on IDA 14 in 2005.
6 The total reduction in IDA reflows due to HIPC for the years 2002 to 2021 is $10.24
billion, with a net present value (NPV) of $5.6 billion. Not all of this cost has yet been
booked. The World Bank has identified sources of funding for some of this. However, its
unfunded HIPC liability is $8.6 billion in nominal terms, about $4.5 billion in NPV terms.

In July 2001, President George W. Bush proposed that up to half the aid the
World Bank provided to poor countries in future years should be in the form of grants
rather than low-cost concessional loans. The President said that such a change in
IDA procedures would help lessen the debt burden on poor countries, provide means
for greater effectiveness and accountability, and put increased emphasis on assistance
for education, health and sanitation. Critics of the proposal say it could lead to
conflict and competition with other international programs, divert attention from
poverty-alleviation and other concerns, and require substantial new contributions by
donors. If the proposed grant program is not funded, they add, it would substantially
undercut the financial basis of IDA and force substantial cuts (or termination) of its
aid program in future years. The British and other European governments have been
particularly critical of the proposed plan. Negotiations to seek a compromise and
resolution of the issue are currently underway. For further discussion, see CRS
Report RL31136, World Bank: IDA Loans or IDA Grants.
Many analysts have sought to determine the likely cost and financial feasibility
of the President’s proposal for IDA grants. The World Bank says the cost would be
about $100 billion, in nominal dollars. Of this, $59 billion was actual lost repayments
and $41 billion was foregone interest income IDA would have received if those
reflows had been invested (until used) through IDA’s cash balance. GAO says the
cost would be $15.6 billion, in net present value (NPV) terms for the whole $100
billion and $9.7 billion in NPV terms for the actual lost reflows.7 It is important to
note that the GAO and World Bank estimates are basically the same. The Bank’s
approach counts the cost as it occurs over the 40 year period. The GAO concept
accounts for the time value of money. In effect, it presumes that the costs identified
by the Bank could be covered if a $9.7 billion fund were established today and the
proceeds in that fund were used to offset future losses in reflows as they occur.
Table 3 (below) provides another estimate, based on the analysis in Table 1, of
the likely cost (in nominal dollars) of lost reflows due to grants.8 According to Table
3, the actual reflows lost if IDA instituted a grant program comparable to that
proposed by the Administration would be about $64 billion. This is comparable in
concept to the World Bank’s figure of $59 billion). Costs are counted only in the
year in which they are incurred. Table 3 does not include any estimate for the
imputed cost of lost income which might be generated by those lost reflows.9

7U.S. General Accounting Office. Developing Countries: Switching Some Multilateral
Loans to Grants Lessens Poor Country Debt Burdens. April 2002. GAO-02-593. The
present author thanks GAO analysts for several helpful suggestions received during his
preparation of this CRS report. .
8 The methodology for Table 3 is the same as that for Table 2. It shows the changes which
would occur in IDA’s bottom line (cash balance) if the lost reflows attributable to a grant
program (comparable to that proposed by the Administration) were added to IDA’s costs.
9 Not shown in Table 3 are the likely consequences (according to the model) if IDA tried to
do HIPC debt relief and a major grant program simultaneously without offsetting
contributions by donors. In that case, IDA’s cash balance would become negative in 2009
and the total would reach minus $10 billion by 2021. Without data on HIPC losses beyond
that point, it is impossible to determine what the balance would be in 2041. It would likely
be more substantial, though, than that shown in Table 3.

The important thing is not the differences among the numbers, which are
relatively small given the fact that they are all projections over 40 years. The
important thing is the fact that a grant program will have costs, either to the donors
(more contributions) or to IDA’s beneficiaries (a smaller program) which will need
to be considered if the program is put into effect.
Table 3. Projected IDA Cash Balance, 2002-2041
With Unfunded Grants
(Billions of U.S. Dollars)
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Lost reflows0.
Balance 11.4011.2510.8610.067.635.464.573.051.791.90
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Lost reflows0.
Balance 0.45-0.78-0.96-2.50-3.35-3.29-4.54-5.53-5.46-6.57
2022 2023 2024 2025 2026 2027 2028 2029 2030 2031
Lost reflows0.660.790.911.051.201.331.471.611.741.88
Balance -7.86-9.09-10.41-11.76-13.11-15.74-18.42-21.08-23.82-26.64
2032 2033 2034 2035 2036 2037 2038 2039 2040 2041
Lost reflows2.032.192.332.502.662.812.993.163.333.52
Balance - 30. 79 - 34. 44 - 37. 47 - 40. 63 - 43. 97 - 4 7. 53 - 5 1. 28 - 55. 21 - 5 9. 34 - 64. 58
Source: Created by CRS using Table 1 data and standard disbursement/reflow formulas.
GAO proposed that, instead of making contributions to defray the cost of grants
as they occur, donors might make advanced contributions to a special fund. The
principal and interest in that fund could be used to defray the future cost of lost
reflows due to grants. GAO said that the cost of grants could be offset if donor
countries increased their contribution to IDA by 1.6% annually over the next 40
years. This is compound interest. Over the course of 40 years, using the financial
model presented in Table 1, donors would contribute $64 billion, a 40% increase
over planned contribution levels.
Table 3 shows that, unless IDA grants are funded with increased contributions
from donors, IDA’s cash balance will dip into the negative in the first year in which
repayments would have been otherwise due for grants made in 2003.10 The growing
negative balance reflects the cumulative amount IDA would need to receive in
donations to keep its aid program at planned levels despite the drop in reflow

10 The World Bank estimated that IDA’s cash balance would become negative (absent any
funding for a grant program comparable to that proposed by the Administration) in 2015.
World Bank. IDA Funding Scheme..., p. 14. The Bank also estimated that, if IDA instituted
an unfunded 10% grant program, the total cost in lost reflows would be about $2 billion and
IDA’s cash balance (though small) would remain positive through 2021. It made no
estimate for the cost or impact on IDA’s cash balance for later years (when, by most
estimates, the bulk of the losses would be incurred.).

income.11 The table assumes that IDA will levy a small commitment fee (here
charged as 1.5% of the total in the year the grant is made) to cover administrative
costs. This approximates on an NPV basis the IDA commitment fee for grants which
the World Bank reportedly has under consideration.) Over time, as the size of IDA’s
outstanding loan balance grows more slowly (because more of its aid takes the form
of grants), IDA will likewise receive less income annually from the service charge
it levies on those loans.
Using data that was not available originally to GAO, CRS did a separate
analysis of the amount that would be needed in additional contributions to offset the
likely costs of an IDA grant program. This found that the donor countries would
need to increase their contributions to IDA at an annual rate of 1.776% during the
next four decades to offset the cost of a major grant program. Again, as with the
GAO calculation, this would be compound interest. On this basis, using the financial
model presented in Table 1, donors would contribute an additional $72 billion to IDA
(a 45% increase over current plans) during the next 40 years. This is somewhat more
than GAO calculated. However, the differences between the two estimates are not
significant due to differences in methodology and the difficulty of estimating costs
over such a long period of time. Were the IDA donors to agree on a contribution
plan of this sort, to offset the future costs of grants, the World Bank – having more
complete data on IDA finances – should be able to determine an annual contribution
rate sufficient to accomplish that goal.
The United States currently contributes about 20% of the new money provided
to IDA during each replenishment period. Given the relative size of the U.S.
economy, it seems unlikely that other countries will agree to increase their
contribution levels in order that the United States may decrease its. Should the U.S.
contribution share remain at that level for the next four decades, it is likely that the
United States will be asked to contribute an extra $13 billion to $14 billion (over and
above the $32 billion it would give if donor contribution levels remained flat in
nominal terms during that period) to cover the cost of IDA grants.
Balancing Multiple Goals
IDA’s capacity to fund a major program of assistance to developing countries
in the coming decades is heavily dependant on the way the World Bank and the IDA
donor countries handle its finances. Under current plans, loan repayments
(“reflows”) would provide much of the money IDA needs to fund its operations
during the next 40 years. Contributions from donors would bear a declining share of
the load. In nominal terms, donors would be asked to contribute the same amounts
annually during each of those years even though the IDA program would be growing
in nominal terms to compensate for anticipated inflation. In real terms, given the

11 In Table 3, IDA’s cash balance is somewhat higher in the first decade than that shown in
Table 1 because IDA would receive an up-front commitment fee from grants made during
that period. Correspondingly, the IDA cash balance in Table 3 is smaller in later years than
that shown in Table 3 because – in addition to lost reflows – IDA would not be receiving
the normal 3/4 of 1% service charge during that period on the disbursed balance for aid
released as grants rather than loans. The commitment fee used for the calculations
embodied in Table 3 is worth considerably less in NPV terms than the usual service fee.

effects of inflation, the actual cost of the IDA program to donors would decline by

2.3% annually during the entire period.

In addition to the goal of reducing the cost of the IDA program to donors, many
people also hope that the cost to the recipient countries may be reduced as well. The
program of HIPC debt forgiveness for poor countries seeks to reduce the amount that
poor countries must pay to IDA each year as a consequence of earlier loans.
Likewise, the proposal for a major IDA grant program would reduce – if not
eliminate – much of the cost that recipient countries must bear as a consequence of
IDA aid.
Ultimately, the goal of making IDA more self-financing through reflows and the
goal of reducing the recipient countries’ debt repayment burdens are incompatible.
Any debt repayments that are forgiven (through HIPC) or eliminated (through grants)
will reduce the amount that IDA receives in future years as reflows to fund its future
aid program. Without additional contributions from donor countries, either to offset
the cost of IDA’s regular assistance program or to offset the cost of HIPC debt
forgiveness or grants, the resources available to fund IDA’s assistance program will
gradually decline. In that case, IDA will have no real alternative but to shrink the
size of its new assistance program in proportion to the amount it is providing its
beneficiary countries through debt relief and grant assistance. In either context, the
donor countries will need to decide whether they wish to reduce the size of the IDA
program in future years or to increase (in real terms) the size of their annual