Revised May 9, 2000

Trends in U.S. Development Aid and the Current Budget Debate
by Isaac Shapiro (1)

Press Release as PDF

Press Release as HTML

Report as PDF

ODC Report:  Declining Aid Spending Harms U.S. Interests

The share of U.S. resources devoted to development, humanitarian or economic aid for other countries has generally fallen since the mid-1960s. The overall decline has been substantial, reducing such spending to exceptionally low levels for the United States. The share of national resources the United States contributes in aid to the world's poorest nations is now far lower than the share that any other industrialized country contributes, and is at one of the lowest levels in the post-World War II era.

The amount of U.S. development, humanitarian and economic aid expenditures increased somewhat in fiscal year 2000. An examination of both the Clinton Administration's fiscal year 2001 budget and the budget that Republican leaders have crafted on Capitol Hill, however, suggests this modest improvement is likely to be short-lived and the long-term decline is likely to continue. Under the Clinton budget, development aid spending would fall again in fiscal year 2001 and drop further in subsequent years. Congress, meanwhile, just passed a budget blueprint that would likely lead to even lower levels of development-aid spending than those the Administration's budget requests.

Under the Administration's budget, expenditures for development aid in fiscal year 2001 would be well below their historical average. In fact, they would be lower, according to two key indicators, than at any time in the half century before 1998.

Table 1
Discretionary Spending on Development Aid

1980s Average

2001, Clinton Proposal

Comment on 2001 Figures

Inflation-adjusted level

$13.1 billion

$10.7 billion

Below average in dollars spent
As a share of the economy



Tied for post-World War II low
As a share of the budget



Except for 1998 and 1999 lowest on record
Source: Center on Budget and Policy Priorities analysis of data from the Office of Management and Budget and U.S. AID

In the 1980s, discretionary spending for development aid averaged 0.20 percent of the economy. The share of the economy dedicated to development aid has fallen by nearly half since then. The decline is even larger when measured over a longer period.

In addition, under the Administration's budget, development aid would decline further in the years after 2001. By fiscal year 2003, U.S. development aid as a share of the economy and as a share of the federal budget would reach the lowest levels on record.

This study also examines data from the Organization for Economic Co-operation and Development (OECD). These data enable U.S. aid expenditures targeted on poor countries to be compared to the efforts of other OECD donor countries, such as the United Kingdom, Germany, France, and Japan.

OECD data on development aid as a share of the economy in each of the 21 countries are available back to 1950. The budget office of the U.S. Agency for International Development tracks U.S. economic assistance back to 1946.(3) The definition of aid that AID uses differs somewhat from the definition the OECD uses and also differs from the definition reflected in the OMB data cited above. Nevertheless, the data from these three sources indicate that under the Administration's budget, U.S. development aid in fiscal year 2001 would be tied for its lowest level since the end of World War II, measured as a share of the economy.(4) Moreover, in subsequent years, the share would fall further, to record lows for the post-World War II period.

These findings also are consistent with a Congressional Budget Office report that tracked foreign aid through 1997, using a definition of aid that includes military assistance. CBO found that "U.S. spending on foreign aid has fluctuated from year to year but has been on a downward path since the 1960s."(5) CBO also reported that "as a percentage of federal outlays, U.S. spending on foreign aid reached its lowest level in the post-World War II era in 1997."

This paper also briefly examines how resources and living standards in the United States compare with those of other nations. On average each person in the United States has 56 times the average annual income of each person in the world's low-income countries. The World Bank estimates that 1.2 billion people in developing or transitional countries live on less than $1 a day.

These trends and international comparisons provide some context for deliberations that will occur in coming months concerning what level of funding should be provided for U.S. development aid in fiscal year 2001. The United States would have to increase its aid commitment substantially either to reach the levels other wealthy nations provide or to return to its own level of commitment in earlier decades. But Congress seems more likely to reduce funding for development aid below the levels the Administration has requested than to raise funding above the Clinton level. While Congressional leaders have not proposed a specific level for development spending, the budget resolution that Congress approved on April 13 includes a significantly lower level of funding for international affairs — the budget category that includes development aid, along with other international spending — than the Administration has proposed. The budget resolution includes six percent less than the Administration proposed for international spending in 2001, and 17 percent less by 2005.

Before proceeding with the body of this paper, at the outset it should be made clear that this paper focuses on the important, but limited, question of the level of assistance the United States devotes to economic aid provided to other countries, especially poor countries. Other important questions concerning development aid — such as the effectiveness of such aid, how its effectiveness can be strengthened, and how the United States might better target its aid — are beyond this paper's scope. Those issues, however, are discussed in a companion brief the Overseas Development Council has produced.


Trends in U.S. Spending Levels

Development aid can be defined and measured in a number of ways. The definition used here begins with the budget category called "International Development and Humanitarian Assistance." This category includes funding for areas such as bilateral development aid, assistance to countries in transition (mainly those that once were part of the former Soviet Union), certain contributions to multilateral institutions that can be used for development aid (such as funds for the development finance division of the World Bank), and as the Child Survival and Disease program and humanitarian aid.  Our focus here is on discretionary spending in these areas.(6)

The definition of development aid used here also includes assistance under the Economic Support Fund. The Economic Support Fund consists of economic assistance the United States provides to foreign countries based on considerations of special economic, political or security needs and U.S. interests. The countries that receive funds from the Economic Support Fund are specifically chosen for their strategic importance, and not all of these funds are targeted on development. As a result, lumping Economic Support Funds in with other development aid somewhat overstates the total amount of such aid. Nevertheless, the economic aid that countries receive from the Economic Support Fund can assist economic development. As a result, this assistance is considered here as part of development aid. This paper does not treat military assistance as part of development aid.

Under the Clinton budget, this figure would begin to drop again, to $10.7 billion in 2001. (Year by year data can be found in Appendix 1.)

Figure 1

Figure 1

Source: Center on Budget and Policy Priorities analysis of Office of Management and Budget data. Figures for 2000-2005 reflect the Administration’s budget proposal.

It bears noting that overall discretionary spending on international affairs — which also includes military aid, contributions to the United Nations and other international organizations, and general State Department operations — would rise in 2001 under the Clinton budget to $23.7 billion, or 0.24 percent of the U.S. economy. That would still be a smaller share of the economy and of total federal spending than in any year from 1962 to 1995. Under the Administration's budget, international affairs spending would again decline after 2001, not only as a share of the economy and federal spending, but also in inflation-adjusted terms. Overall, by 2004 international affairs spending as a share of the economy would be at its lowest level on record. (As noted, these OMB data go back to 1962.)

In coming months, Congress may set fiscal year 2001 spending for development aid at a level lower than the Administration has proposed. The budget resolution that Congress just approved includes a fiscal year 2001 discretionary spending figure for international affairs that is $1.4 billion below the Administration's proposed level. By 2005, the figure in the budget resolution adopted is $3.9 billion, or 17 percent, below the Administration's proposed level. Although the budget resolution does not indicate how much of the international affairs budget should be devoted to development aid and merely establishes targets for binding appropriations legislation to be passed later this year, Congress appears to be charting a course that would direct significantly fewer resources to this area than would the Administration's budget, which itself reflects development aid levels that are low in historical terms.


U.S. Compared to Other OECD Countries

The Organization for Economic Co-operation and Development (OECD) is the authoritative source for information comparing the extent to which various industrialized nations devote resources to development assistance. The OECD examines the amount of development assistance provided each year by the members of the Development Assistance Committee, which is comprised of 21 countries, including the United States, Canada, France, Germany, Japan, and the United Kingdom.(8)

The measure of assistance to which the OECD accords most attention is "Official Development Assistance" (ODA), which consists of development aid to the world's poorest countries, as identified by the United Nations.(9) The OECD recently released data on this measure through 1998.(10)

These data show that the share of Official Development Assistance world-wide that the United States contributes has dropped dramatically over the past four decades. (In the following data, the OECD practice of combining and averaging two years to help smooth out yearly fluctuations is often employed.)

Table 2
U.S. Official Development Assistance Compared to Other Countries

United States

Typical Donor Country*

ODA as a share of the economy



ODA per person in the donor country



* Typical Donor Country figures represent the median levels of the 20 countries other than the United States examined by the OECD
Source: Development Co-Operation, 1999 Report, OECD, 2000

Another way that the OECD measures national commitment to Official Development Assistance is by examining the amount of assistance extended on a per capita (that is, per person) basis. For each person in the United States, the U.S. contributed an average of $29 per year to ODA in the 1997-1998 period. The median country among the 21 nations the OECD examined contributed $70 on a per capita basis. (OECD expresses these figures in 1997 dollars.) On the high side, Denmark contributed $316 per person to ODA. Of the 21 countries, only Portugal — a far poorer country than the United States — contributed a smaller amount of funds on a per capita basis.

The OECD also tracks "Official Aid." This includes the same types of economic aid as ODA, but differs from ODA in that the countries receiving this aid are considered better off than the countries receiving ODA.(14) The amount of Official Aid extended world-wide is much lower than the amount of ODA provided. Partly because U.S. economic aid is less well-targeted on poor countries than the aid that other industrialized nations provide, the United States ranks above average in the amount of Official Aid it provides, measured as a share of the economy.(15)

When Official Assistance and ODA are combined, however, the United States continues to compare unfavorably. In 1997-1998, approximately 0.13 percent of U.S. GNP went to Official Aid and ODA combined. This was the lowest share of any of the 21 countries. In fact, it was well below half of the average share of GNP (0.33 percent) that the other countries contributed in Official Aid and ODA combined.


How do Resources and Living Standards in the United States Compare with Those of Other Nations?

In assessing the relative level of resources the United States devotes to development aid, it is useful to examine the wealth of the United States relative to that of other countries, as well as how living standards here compare with those in countries to which we might provide assistance. World Bank data are particularly useful in this regard.(16)

In South Asia alone (which includes India), an estimated 522 million people — 40 percent of the population — lived on less than $1 a day in 1998. In Sub-Saharan Africa, 291 million people — 46 percent of the population — did. The 291 million people in Sub-Saharan Africa who were this poor constitute a larger number of people than the entire U.S. population.

United Nations data provide some sense of the living conditions associated with such low levels of income.(18)

In the developing world, reflecting the inadequate food intake and poor health conditions, an estimated two or every five children are stunted (i.e., have significantly low height for their age), and one in three is underweight.



The trends in U.S. development aid are unmistakable. The U.S. commitment to development aid has fallen significantly over time and is quite low now compared to the levels that other industrialized nations provide. Both the Administration's budget and the emerging Congressional budget would continue this pattern, allowing development assistance funding to decline further. In the first years of the 21st century, the United States is likely to continue contributing a much-smaller fraction of its resources to economic aid to poor countries than any other industrialized nation on the globe.

Appendix 1. Development Aid Trends, 1962 - 2005


Aid in 2000 Dollars 
(in billions)

Aid as % of GDP

Aid as % of Budget Outlays

















































































































































































*Data for 2000 through 2005 are from the Administration's budget proposal
Source: Center on Budget and Policy Priorities analysis of Office of Management and Budget data


Appendix 2. Comparing Pre-1992 Data to Data from 1992 On

Although the Office of Management and Budget provides annual data on discretionary spending for fiscal year 1962 onwards, the time series is not entirely consistent. The inconsistency reflects the difference in the budgetary treatment of loans prior to 1992 and from 1992 on.

Prior to 1992, loans were recorded in the budget on a cash basis; that is, the amount associated with making a loan equaled the cash disbursed. Similarly, principal repayments and interest paid on the loan were recorded on a cash basis as "negative spending," in other words, as a reduction in spending. As a result of the Federal Credit Reform Act of 1990, since 1992 loans have been treated on a subsidy basis; that is, the amount of spending associated with a loan equals the estimated amount of the subsidy provided by a loan. Principal repayments and interest paid on the loan are recorded in "non-budgetary" accounts and thus do not affect the spending recorded in the budget. OMB has found it "impossible to convert the pre-1992 loans to a credit reform basis," and thereby to construct a consistent time series.

As an example of the effects of these differences in treatment, take the hypothetical case of a loan of $1 million that involved a subsidy of $200,000.

The net effect on the measurement of foreign economic aid is difficult to determine. It depends on the nature of the loans issued, the timing of the repayments, and changes in the size of the loans over time. That is, we know the time series is not entirely consistent, but it is difficult to ascertain whether in comparing particular years this inconsistency leads to an overstatement or an understatement of the fall in economic aid over time.

Nevertheless, it is possible to state with confidence that foreign economic aid has declined since the 1960s, and is now at or near the lowest levels on record as a share of the economy and government outlays, which was the conclusion reached by the Congressional Budget Office study cited on page three of the report. First, although the proportion of economic aid that consists of loans is now lower than it was, the data that are available suggest that loans were not the majority of foreign economic aid prior to 1992. U.S. AID data for the period from 1962 to 1993 indicate that 21 percent of their definition of foreign economic assistance consisted of loans, the other 79 percent consisted of grants. (AID did not break these data out on an annual basis.)

Second, the differences between now and the 1960s and 1970s are dramatic (because of the drop in loans during the 1980s, there is little reason to be concerned that comparing current spending to spending in the 1980s leads to an overstatement in the fall of foreign economic aid), suggesting that an inconsistency of difficult-to-determine direction is unlikely to affect the basic trend. From 1962 to 1969 according to the definition of foreign economic aid used in this paper, it equaled between 0.27 percent and 0.58 percent of the economy; in the 1970s, foreign economic aid constituted between 0.18 percent and 0.25 percent of the economy; under the Clinton budget it would equal 0.11 percent of the economy. The differences between now and then are so substantial that they are extremely likely to remain even if one were able to account for the loans on a consistent basis.

End Notes:

1. The author would like to thank James Horney, Robert Greenstein, Stefan Falk, Betty Hitchcock, Ric Loll, and Michael Garrett of the Center for their comments, help and advice on this paper, as well as ex-Center employee Abja Midha for conducting the initial research for the report. Thanks also to Jeffrey Colman and David Weiner for their reviews of this paper and their suggestions.

2. The definition of development aid used here includes all discretionary spending within subfunction 151 of the federal budget (subfunction 151 is international development and humanitarian assistance), as well as the Economic Support Fund. The Economic Support Fund is part of subfunction 152, international security assistance.

3. U.S. Overseas Loans and Grants: and Assistance from International Organizations (Statistical Annex I to the Annual Development Coordination Committee Report to Congress), Office of Budget in the Bureau of Management of the U.S. Agency for International Development. The data in this publication are not always provided for each individual year. For example, the 1946-48 figures are lumped together into one number.

4. Based on the U.S. AID data, U.S. "Economic Assistance" (the category AID uses that is somewhat analogous to the definition of development aid this paper uses) averaged 1.75 percent of GDP from 1946 to 1948. From 1949-1952, it averaged 1.53 percent of GDP, while from 1953-1961, it averaged 0.60 percent of GDP. Over the 1962-1993 period, it averaged 0.27 percent of GDP. In 1997, the latest year the U.S. AID data cover, Economic Assistance equaled 0.11 percent of GDP.

Switching to the data cited in this paper that cover recent years and are based on data from OMB, discretionary spending on development aid equaled 0.12 percent of GDP in 1997. In both 1998 and 1999, it equaled 0.11 percent of GDP, the lowest share on record. Under the Administration's budget, development aid would again equal 0.11 percent of GDP in 2001 and edge down to 0.10 percent of GDP in 2002.

5. The Role of Foreign Aid in Development, Eric J. Labs, Congressional Budget Office, May 1997.

6. Mandatory spending in these areas consists primarily of inflows of funds into liquidating accounts. In 2000, the net figure for mandatory spending on development aid is minus $1.4 billion (i.e., there will be an in-flow of $1.4 billion). If the figures for discretionary and mandatory spending are combined, spending on foreign development aid is lower than the figures cited in this paper — expenditures in fiscal 2000 are estimated to be $9.6 billion, rather than the $11 billion figure used here. This is because the $11.1 billion in discretionary spending will be offset by $1.4 billion in mandatory receipts, for a net combined level of $9.6 billion.

7. According to data from OMB, in 1962, the earliest year for which there data are available, development aid equaled 0.58 percent of the economy, about five times its current level. Those figures are not strictly comparable, however, because of the difference in the treatment of loans prior to 1992 and since 1992. See Appendix 2 for an explanation of this issue.

8. The other countries on the Committee are Australia, Austria, Belgium, Denmark, Finland, Ireland, Italy, Luxembourg, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, and Switzerland.

9. Specifically, the OECD's measure of ODA consists of "Grants or loans to countries and territories on Part 1 of the DAC List of Aid Recipients (developing countries) which are: undertaken by the official sector; with promotion of economic development and welfare as the main objective; at concessional financial terms (if a loan, having a GRANT ELEMENT of at least 25 per cent)." This definition does not encompass grants, loans, or credits for military purposes.

10. Development Co-operation, 1999 Report, OECD, 2000.

11. A very small part of this percentage decline reflects the addition of new donor countries to the list for which OECD compiles data. For example, in the 1977-1978 period, the OECD did not examine contributions from Luxembourg, Portugal or Spain. These countries are included in the 1987-88 and 1997-98 data. These countries account for only a tiny fraction of ODA. Had these countries not been included in the OECD list in recent years, the U.S. share of total ODA would have been 15.9 percent in 1997-1998 instead of 15.6 percent when these countries are included.

12. The U.S. "Gross National Product" (used by OECD) and the U.S. "Gross Domestic Product" (used earlier in this paper and in most U.S. data) are virtually identical. In 1998, U.S. GNP and GDP differed by only one-tenth of one percent.

13. The United States dedicated 0.21 percent of GDP to official development aid in the 1987-1988 period.

14. Official Aid (OA) goes to "Part II" countries, which are those considered to be in transition, including Central and Eastern European Countries and the New Independent States of the former Soviet Union, as well as more advanced developing countries and territories such as Israel.

15. In the 1997-98 period, the United States contributed 0.03 percent of GDP to Part 2 countries. The average in the other donor countries that the OECD examined was 0.02 percent.

16. Entering the 21st Century: World Development Report 1999/2000, The World Bank, August 1999 and Poverty Trends and Voices of the Poor, The World Bank Group, December 2, 1999.

17. The number of people in developing and transitional countries living on less than $1 a day was about the same in 1998 as it had been in 1987. The share of the population with such income levels did fall, from 28 percent of the population of such countries in 1987 to 24 percent in 1998.

18. Human Development Report 1999, United Nations Development Program, 1999.