_ . ..-.._ .._. -..--_.._...__..ll”“. ..-.--.“_. I. GAO -.._ ^..-._” ,~._._.I~....^... .-_..-. _.I”..*._ l.“.l”.-.._.-.” _...-.- -.- .--..--.__~-_-I__- -- Stk[)i Plklf)t’l’ I!)!)() F’OREIGN ASSISTANCE Use of Host Country- Owned Local Currencies i-3 v) IllIIllllII 142506 General Accounting Office unless specifically approved by the Office of Congressional Relations. GAO/NSIAI)-!M-2 1OIM United States GAO General Accounting Office Washington, D.C. 20648 National Security and International Affairs Division B-238869 September 26,199O The Honorable Dante B, Fascell Chairman, Committee on Foreign Affairs House of Representatives Dear Mr. Chairman: As requested, we are providing information about local currencies gen- erated as a result of U.S. assistance to foreign countries. This report summarizes the information we provided to your staff on January 10, 1990, and includes some updated information. Specifically;the report addresses (1) ownership of the local currencies, (2) whether use of the currencies was consistent with U.S. assistance objectives, (3) the accountability requirements for local currency use, and (4) the potential for local currency generation and its use to affect host country econo- mies. As part of our review, we conducted field work in Tunisia and Zaire. These countries may not be representative of other countries receiving U.S. assistance, and therefore, conclusions on the use of local currency in these two countries cannot be generalized to other countries. Local currencies are generated primarily through financial or com- Background modity assistance provided under Public Law 480, the Economic Sup- port Fund, and the Development Fund for Africa. For example, under certain programs, the United States provides commodities to recipient governments. The governments sell the commodities, usually in the pri- vate sector, and the proceeds are local currencies that the government would not otherwise have had. In other cases, the United States pro- vides US. dollars as assistance, and aid agreements may require the recipient government to make available local currencies in an amount up to the value of the U.S. dollars. In fiscal year 1988, U.S. assistance- generated local currencies equaled $1.4 billion, about 48 percent of which was attributable to Public Law 480 programs, about 48 percent to Economic Support Fund financial contributions, and the remainder to the Development Fund for Africa. Data on 1989 generations was not yet available at the time of our audit work. The Congress defines the allowable uses for local currencies generated under each of these assistance programs. In general, most currencies are to be used for economic development purposes, and in support of U.S. development objectives. As a result of legislation and assistance agree- ments, the Agency for International Development (AID) and host country Page 1 GAO/NSIAD-SO-21OBR Foreign Assistance B-238869 rl officials jointly program, or negotiate and mutually agree on how these local currencies are used. AID is responsible for ensuring that the local currencies are used for the agreed upon purposes. Local currencies are used for a wide variety of activities. They support development projects or programs managed by AID or the host country. They are also used for (1) general budget support in the host country, (2) the development budget of a particular ministry or sector, or (3) AID mission operating expenses. In some cases, local currencies can be used only after the host country has made certain policy reforms as agreed upon with the United States. When the United States provides commodities or dollars as assistance to Results in Brief recipient countries, the local currencies that are generated from this assistance are owned by the host country. The currencies are not an added resource to the host country economy-the added resource is the initial U.S. assistance provided in cash or commodities. However, this transfer of local currencies from the private to the public sector allows participating governments to retain control over the additional spending power generated by the initial assistance. In Tunisia and Zaire, local currencies were programmed in fiscal year 1988 for uses generally consistent with U.S. development objectives, and we believe the currencies were generally used for the agreed upon projects and programs. However, worldwide, even when local currencies were programmed to support U.S. development objectives, AID’S over- sight of local currency use has not always been adequate to reasonably ensure that the currencies were consistently used for the agreed upon purposes. AID is considering improved oversight guidance that will prop- erly focus on the responsibilities and financial management capabilities of the host government. Generating and spending local currencies can contribute to inflation or deflation in the recipient country. However, in most countries, it is unlikely that, in 1988, generating and spending local currencies resulting from U.S. assistance had a significant macroeconomic effect. Also, in most countries, accumulations of unspent local currencies are not large enough to have a significant future macroeconomic effect; although we identified instances where the potential for a macroeconomic effect exists. Page 2 GAO/NSIAD-BO-210BR Foreign Assistance B-238869 We believe that because the United States structures its assistance as a Host Countries Own grant or a concessional sale, it signals its intent that the host country the Currencies will own the currency generated from the assistance. Nevertheless, AID is responsible for ensuring that the currencies are used for the agreed upon purposes. officials participate in decisions on how host country development Local Currency AID resources are used as a result of joint programming requirements. AID Programming in officials told us that they hold discussions with host country officials on Tunisia and Zaire the proposed uses for local currencies throughout the year, not just during formal agreement negotiations, However, it is not always pos- Supported sible to conclude that these discussions influence or change host country Development decisions and resource allocations. For example, Tunisian government officials said that, even in the absence of joint programming, they would Objectives have used the local currencies generated from food aid in the same way -in support of public works employment programs. Also, Tunisian officials did not make those policy changes suggested by AID mission officials that they believed were not in their best interest, even though $3.0 million in local currency remained unused as a result. In Tunisia, local currencies equal to $31.6 million were spent in fiscal year 1988. The currencies funded government public works projects designed to promote agricultural and rural development and provide temporary employment to counteract the negative effects of Tunisia’s structural adjustment program. Tunisian officials place a high priority on these public works projects, and local currency support for the projects fostered political good will between Tunisia and the United States. Local currencies were also programmed to support a private sector program and to speed policy reform consistent with U.S. objec- tives that promote open market private sector activities. In Zaire, local currencies equal to $24.3 million were spent in fiscal year 1988. More than 70 percent of the local currencies programmed in 1988 supported AID mission projects, primarily agricultural, rural develop- ment, and health projects. Local currencies were also used for AID mis- sion general operating expenses and mission costs associated with supporting the development projects. Finally, local currencies were programmed for budget support for the Zaire planning ministry and two development ministries, and to support a special unit formed to manage the local currencies generated by US. aid and that of six other donors. Joint Zaire/AID accounting and monitoring requirements and AID tech- nical assistance are intended to help reduce fraud and abuse in Zaire. Page 3 GAO/NSIAD-SO-2lOBR Foreign Adratance B-238869 is responsible for providing reasonable assurance that the local cur- Accountability for AID rencies are used for the agreed upon purposes, even though the curren- Local Currency cies are owned by the host government. Our previous reports, listed on the last page of this report, and AID Office of the Inspector General (IG) audits have found that AID's monitoring was insufficient to provide rea- sonable assurance that the currencies were properly used. AID recognizes that the local currencies are vulnerable to waste, fraud, and abuse. A 1989 survey of AID missions identified key weaknesses in local currency oversight. As of April 1990, AID'S Office of Financial Man- agement had proposed stronger and clearer guidelines to address these weaknesses, including requiring formal and standard financial assess- ments and audits of host country agencies that use local currencies. We support this approach, because it focuses on improving the financial management systems of these host country agencies. The guidelines also touch on recommendations made in our previous reports. Under certain conditions, local currency generation and spending may Potential for have a macroeconomic effect in the host country. For example, when Macroeconomic Impact commodities are sold in the private sector and the currencies are depos- Is Small ited in government accounts, the money supply contracts. This results in a deflationary effect. When the currencies are spent, the money supply expands, which may result in an inflationary effect. However, if the generation and expenditure occur at about the same time, the infla- tionary and deflationary effects are not likely to occur. When the cur- rencies are not spent at about the same time as they are generated, the larger the volume of local currencies involved, compared to the host country money supply, the greater the potential that the local currency activity can have a macroeconomic effect. We believe it unlikely that, in 1988, generating and spending local cur- rencies resulting from U.S. assistance had an adverse macroeconomic effect in most countries. An analysis of the size of local currency pro- grams generated from U.S. assistance and the timing of generations and expenditures in 1988 among 47 countries indicated that, in general, cur- rencies were spent soon after they were generated and the volumes of local currencies were small compared to the host country money supply. However, in six countries, the currencies were not spent soon after they were generated and/or the volume of local currencies was large enough to suggest that local currency activity could have had a macroeconomic effect. Appendix I provides more details about the generation and use of local currencies. Page4 GAO/NSIAD90-210B&Fo~~Asslstance B-236660 and the Department of Agriculture agreed with the findings and con- Agency Comments AID clusions of this report in written comments, which are included in appendixes II and III. Agriculture suggested minor modifications to some of the report language which were incorporated where appropriate. As agreed with your Office, we plan no further distribution of this report until 30 days after its issue date. At that time we will send copies to the Administrator of AID, the Secretary of Agriculture, and appro- priate congressional committees. We will also make copies available to others upon request. If you have any questions concerning this report, please call me at 275 6790. Major contributors to this report are listed in appendix IV. Sincerely yours, Harold J. Johnson Director, Foreign Economic Assistance Issues Page 6 r Conknts Appendix I 8 Programming and Use Background 8 Scope and Methodology 11 of Local Currencies Who Owns the Local Currencies? 13 Why Generate and Program Local Currencies? 16 How Are Local Currencies Used Worldwide? 18 Was Local Currency Use in Tunisia and Zaire Consistent 20 With U.S. Assistance Objectives? Local Currency Accountability Requirements 30 Did Local Currency Use Impact on Host Country 33 Economies? Appendix II 39 Comments From the Agency for International Development Appendix III Comments From the GAO Comments Department of Agriculture Appendix IV 44 Major Contributors to This Briefing Report Related GAO Products 48 Tables Table 1.1: 1988 Local Currency Generation, Expenditure, 36 3 and Pipeline as a Percent of Broad Money Supply i Table 1.2: 1988 Net Local Currency Expenditure as a 37 Percent of Broad Money Supply Page 6 GAO/NSIAIMO%lOBR Foreign Assistance Contenta - Figures Figure I. 1: Background: Local Currency Generation and 8 Sources Figure 1.2: Fiscal Year 1988 Local Currency Generations 9 Figure 1.3: Scope and Methodology 11 Figure 1.4: Who Owns the Currencies? 13 Figure 1.5: Why Generate and Jointly Program Local 15 Currencies? Figure 1.6: How Are Local Currencies Used Worldwide? 18 Figure 1.7: Worldwide Local Currency Use in 1989 19 Figure 1.8: Tunisia: Was Use Consistent With Assistance 21 Objectives? Figure 1.9: 1988 Local Currency Expenditures in Tunisia 23 Figure I. 10: Zaire: Was Use Consistent With Assistance 26 Objectives? Figure I. 11: 1988 Local Currency Expenditures in Zaire 27 Figure I. 12: Accountability Requirements 30 Figure 1.13: Local Currency Factors Affect Host Country 33 Economies Figure I. 14: Did Local Currency Use Impact on Host 36 Country Economies? Abbreviations AID Agency for International Development DFA Development Fund for Africa EJF Economic Support Fund IG Office of the Inspector General Page 7 GAO/NSIAD-BO-210BR Foreign Arxdstance $ Appendix I progr amming and Use of Local Currencies Flaure 1.1 w Background: Local Currency Generation and Sources Local currencies are generated when the United States provides *Food or other commodities, which the recipient sells *Dollars, and requires recipient to match with local currencies A Local currencies are generated through U.S. foreign assistance and food Background aid programs. The United States, for example, may provide agricultural or other commodities to a developing country through a grant or conces- sional sale. Local currencies are generated when the recipient govern- ment sells these commodities. Also, when the United States provides US. dollars as assistance, AID can require the recipient government to make local currencies available for mutually agreed upon purposes in an amount up to the value of the dollars. Page 8 GAO/NSUD-90-210BR Foreign Assistance Appendix I Programming and Use of Local Currenciee Figure 1.2: Flrcrl Vear 1999 Local Currency Oenerations ti ESF ($6!3.2 million) 3.9% DFA ($52.7 million) 47.9?6 Food Aid ($657.1 million) ; \ I _A Note: Dollar equivalents represented were calculated with average annual exchange rates. Local currencies are generated from three sources-the Economic Sup- port Fund @SF), the Development Fund for Africa (DFA), and food aid programs, as shown in figure 1.2. In fiscal year 1988, local currencies equivalent to $1.4 billion were generated from U.S. assistance. Complete data on fiscal year 1989 was not yet available during our audit work. The Foreign Assistance Act of 1961, as amended, authorizes ESF pay- ments to countries of particular security and political importance to the United States. Through FSF, local currencies may be generated from the U.S. assistance when the United States provides dollars as cash grants, or commodities through commodity import programs to a recipient country. At least 60 percent of the local currencies generated under commodity import programs are to be used to meet the development objectives targeted under the Foreign Assistance Act. In fiscal year 1988, ESF generated an equivalent of $653 million in local currency, or 48 percent of total generations. The DFA was first funded in fiscal year 1988. Similar to ESF, the United States provides dollars or commodities, which generate local currencies. All currencies must be used for development activities consistent with the Foreign Assistance Act and for necessary U.S. government adminis- trative requirements. In fiscal year 1988, DFA generated an equivalent of $53 million in local currency, or 4 percent of the total generations. Page 9 GAO/NSLAD-90-210BR Foreign Assistance Appendix I Programming and Use of Local Currencies The United States provides food aid to developing countries under the Agricultural Trade Development and Assistance Act of 1954, as amended, commonly referred to as Public Law 480. Under title I of Public Law 480, countries acquire agricultural commodities through long-term, low-interest purchases. They sell the commodities in the pri- vate sector to generate local currency, but must pay for most of the purchases eventually in dollars. Title III is similar to title I, but forgives the concessional credit repayment when local currencies are used for specific development purposes. Under title II, the United States donates food for humanitarian purposes, and some of this food is also sold by recipient governments for local currency. Local currencies generated from these titles must be used for economic development to benefit the poor in areas such as agricultural and rural development and nutrition. Finally, under section 416(b) of the Agricultural Act of 1949, the United States donates surplus agricultural commodities to developing countries. Proceeds from the sale of these commodities are to be used for develop- ment programs or to benefit the needy. In fiscal year 1988, Public Law 480 and section 416(b) of the Agricultural Act of 1949 generated an equivalent of $657 million in local currencies, or 48 percent of the total generations. Page 10 GAO/NSIAIMO-210BR Foreign Assistance Appendix I Programnhq and Use of Local Currencies Fiaure 1.3 GAQ Scope and Methodology @Workdone in Washington, Tunisia, and Zaire @Focusedon host country- owned local currencies l Limited analysis to 1988 currency uses *Did not conduct an account- ability and control audit We conducted our review in Washington, D.C., and in Tunisia and Zaire. Scopeand In Washington, D.C., we interviewed officials from AID, AID'S IG, and the Methodology Departments of State and Agriculture. We reviewed evaluations and studies of the major assistance programs that generate local currencies. We focused solely on local currencies generated from U.S. assistance and owned by the host country. Unless otherwise specified, all references to local currencies in this report refer to these U.S.-generated currencies. Our field work was performed in Tunisia and Zaire because these coun- tries generate local currencies from at least two types of US. assistance, and differ substantially in their use of local currencies. We interviewed Page 11 GAO/NSLAD-SO-2lOBR Foreign Assistance Appendix I Frogramndng and Use of Local Currencies AID mission and US. Embassy officials, United Nations Development Program officials, representatives from private accounting firms, and host country officials from the Ministries of Planning, Finance, Foreign Affairs, Treasury, Central Bank, and Agriculture. We reviewed project documents, evaluations, assistance agreements, planning documents, and International Monetary Fund and World Bank papers, and visited project sites in Zaire. We focused on fiscal year 1988 because complete data for 1989, including macroeconomic information, was not yet avail- able during our audit work. To determine the potential macroeconomic effect of generating and spending local currencies associated with U.S. assistance, we analyzed the volume of local currency activity relative to host country money supply and the timing of expenditures and generations worldwide. We used local currency data on fiscal year 1988 actual expenditures, as reported in 1991 Annual Budget Submissions for 66 countries with local currency programs. We used economic data extracted primarily from the International Monetary Fund’s International Financial Statistics. Adequate economic data was not available for 8 of the 66 countries. In addition, we discussed host country monetary policy with government officials in Tunisia and Zaire. We discussed our findings with AID and International Monetary Fund economists. We were limited in performing our work. We could not perform a world- wide analysis of local currency use because local currency programming decisions and reporting and monitoring requirements are decentralized and the needed data was not centrally available. We discussed accounta- bility issues with AID program and IG officials and reviewed previous audits and reports, but we did not conduct an audit of local currency controls and accounting systems in Zaire and Tunisia. While the case studies in Tunisia and Zaire provide some examples of how local curren- cies are programmed and of the advantages and disadvantages of the current system, the results cannot be generalized to other countries. We did not assess the effectiveness of the programs supported by local currencies. Our review was performed from August 1989 to April 1990 in accor- dance with generally accepted government auditing standards. AID and the Department of Agriculture provided written comments on a draft of this report, which are included in appendixes II and III. Page 12 GAO/NSIAD-BO-210BR Foreign Assistance Appendix I Programmhg and Uee of Local Cum&m G&I Who Owns the Currencies? l Host countries own the currencies generated from U.S. assistance l Local currencies must be jointly programmed @AID must ensure that local currencies are used for the agreed purposes When the United States provides dollars or commodities as part of its Who Owns the Local assistance program, the local currencies generated from this assistance Currencies? are owned by the host governments. We believe that because the United States structures its assistance as a grant or a concessional sale, it manifests its intent that the currencies generated from the assistance will be owned by the host government. Nevertheless, as a result of program legislation and assistance agree- ments, AID and the host country jointly program, or negotiate and mutu- ally agree on how the currencies will be used. AID is responsible for Page 13 GAO/NSIAMO-210BR Foreign Assistance Appendix I Programming and Use of Local Currencies ensuring that the currencies are used for the agreed upon purposes. Because each country’s development needs and resources differ, AID has delegated responsibility to negotiate and monitor currency uses to its overseas missions. An interagency committee in Washington, DC., reviews and approves proposed Public Law 480 agreements and local currency uses, but overall, mission officials have much flexibility in selecting and monitoring projects and programs using local currency. When the host country and AID mission agree that local currencies can be managed by the AID mission, the currencies are placed in a trust account and are used primarily for mission administrative costs. Because the currencies are held in trust, AID missions must report to the host government on the uses of the trust funds. Currencies used for mis- sion administrative costs are primarily generated through ESF or DFA. Currencies generated through Public Law 480 must be used for develop- ment purposes. Page 14 GAO/NSIAD-BO-21OBRForeign Assistance Appendix I Progmmmlng and Use of Local Currencies Flgure 1.5 w Why Generate and Jointly Program Local Currencies? l Generation generally provides AID and/or host country with resources for development l Joint programming provides AID with opportunities for input into host country budget allocations l Joint programming fosters dialogue Local currencies are not an added resource to the host country Why Generate and economy-the added resource comes with the initial U.S. assistance pro- Program Local vided, whether cash or commodities. However, the local currencies gen- Currencies? erated through this assistance represent a transfer of existing resources, in this case, host country currencies, from the private to the public sector, and allow participating governments to retain control over the additional spending power generated by the initial assistance. For Y example, when the host country sells U.S.-provided commodities in the private sector, the government obtains local currencies it would not otherwise have had. The currencies provide host country officials with Page 16 GAO/NSIAlMO-210BR Foreign Assistance Appendix I Programming and Use of Local Currencies financial resources to acquire locally available goods and services to support programs and projects. Host government and U.S. officials in Zaire and Tunisia view host country-owned local currency as a valuable developmental resource. Joint programming is designed to provide AID with an opportunity to influence host country development decisions by negotiating and mutu- ally agreeing on local currency use with host government officials. Some AID officials told us that this fosters an ongoing dialogue, because they hold discussions with host country officials on the proposed uses for local currencies throughout the year, not just during formal agreement negotiations. In practice, however, we found that joint programming may or may not alter the host government’s spending decisions. In Zaire, for example, both AID and some Zaire officials said that joint programming helped to ensure that development projects were funded, and that, in the absence of joint programming, they had no assurance that local currencies would be programmed for development purposes. In Tunisia, on the other hand, we concluded that these discussions were not the deciding factor in host country decisions or resource allocations. Tunisian government officials said that even in the absence of joint pro- gramming, they would have used the local currencies generated from food aid in the same way because the projects funded by local currencies are very important to the government. However, we were told that even though the projects would have been funded, available local currencies allowed Tunisia to expand the size of the projects. Also, in Tunisia we learned that joint programming for policy reform does not always lead to policy change. In 1988, Tunisia did not make policy changes suggested by AID mission officials because Tunisian offi- cials felt these changes were not in their country’s best interest at the time. This resulted in the equivalent of $3.0 million of local currencies left unprogrammed. Some U.S. and host country officials, and others studying foreign assis- tance programs, have argued that the local currency generated from U.S. assistance should not be jointly programmed and that AID should not then be responsible for oversight of its use, for various reasons. As previously stated, joint programming may not alter the spending deci- sions of the host country, and in some cases, AID and the host country Page 16 GAO/NSIAD90-210BR Foreign Assistance Appendix I Prograndng and Use of Lmal Currenclea cannot agree on uses for the currencies until after significant and pro- tracted negotiations. Also, because the currencies are owned by the host country and are not an added resource to the host country economy, some argue that the country should be permitted to dispose of the cur- rencies as it sees fit. They stress that under Public Law 480 title I food aid, countries must now repay the United States for the agricultural commodities received, although at a concessional rate, and that addi- tional conditions on the assistance are excessive. Finally, when countries are participating in an economic stabilization program, using local cur- rencies for new economic development projects or programs could run counter to the common stabilization requirements of austerity and fiscal restraint. Joint programming fosters dialogue between host country and U.S. offi- cials and helps to ensure that those local currencies required by legisla- tion to be used for economic development purposes are indeed programmed for those purposes. By working with the host government, AID can provide guidance and assistance in determining the best use of the resources. Page 17 GAO/NSIAD-SO-210BR Foreign AssM.ance Appendix I Programming and Use of Local Currencies Figure 1.6 w How Are Local Currencies Used Worldwide? l AID development projects l Host country development projects/programs l Host country budget support l Trust funds and other l Policy reform Local currencies may be used for a wide range of activities. They may be How Are Local programmed in direct support of AID or host country development Currencies Used projects or programs. Local currencies may be used to support the host Worldwide? country’s general budget or the development budget of a particular min- istry or sector. Some local currencies are managed by the AID mission as a trust fund, and are usually used for mission operating expenses.In some cases, local currencies are released only after the host country has ” made certain policy reforms as agreed upon with the United States. Page 18 Appendix I From and Use of Local Currendee The appropriateness of using local currencies not earmarked for devel- opment for host country budget support has been recently debated. Some argue that local currencies should be used solely for economic development purposes, rather than to support the recurring expenses of host governments. In addition, when local currencies are used for host country budget support, it is often difficult for AID to ensure proper accountability for the use of the funds because host government accounting and control systems may not be adequate. On the other hand, some argue that when countries are participating in an economic stabili- zation or restructuring program, it may be more appropriate to use local currencies for budget support than for new economic development projects or programs that could undermine efforts to maintain austerity and fiscal restraint. Also, budget support may be appropriate when a country has a sound development plan consistent with U.S. objectives and allocates its resources accordingly. Figure 1.7: Worldwide Local Currency Use In 1989 7 Eer (Drought Relltaf. Monitoring, etc.) 24Oh - - AID PmjecWPrograms Host Country ProjecWPrugrams Page 19 GAO/NSIAMO-2lOBB Foreign Assistance Appendix I Pcogramndng and Use of Lo& Currendee Figure I.7 shows how local currencies have been used worldwide, as reported to AID'S Office of Financial Management in a 1989 survey. Data was not readily available to assess, worldwide, the extent to which these uses were consistent with U.S. assistance objectives. Such determina- tions, because of the decentralized decision-making and monitoring aspects of the program, can be made only at individual missions, Our work in Tunisia and Zaire indicated that the local currencies were Was Local Currency programmed to be used for purposes consistent with U.S. assistance Use in Tunisia and objectives. Also, based on our field work, AID IG audits, and independent Zaire Consistent With pgro ram evaluations, we believe that the local currencies expended in fiscal year 1988 generally were used for the agreed upon projects and US. Assistance programs, However, we did not conduct an accountability and control Objectives? audit to determine the level of assurance that local currencies in those countries were used for the specified purposes. Also, when local curren- cies are used for budget support in Zaire, the currencies are not moni- tored beyond their disbursement to the proper government account. Page 20 GAO/NSIAD9O%1OBR Foreign AssMance Progmmming and Use of Lwal Currendes Figure I.8 GAQ Tunisia: Was Use Consistent With Assistance Objectives? l Generally consistent l Currencies were used to: *Promote agricultural and rural development @Cushion negative impact of ’ structural adjustment @Fosterpolitical good will *Promote export development @Speedpolicy reform/budget support Local Currency Use in In fiscal year 1988, local currencies helped reduce the negative impact Tunisia on employment from Tunisia’s economic structural adjustment program and the effects of severe drought conditions. These currencies also sup- ported rural public works projects designed to increase agricultural out- puts and promote rural development, supported the development of livestock forage reserves, fostered political goodwill, and hastened policy reform. However, $3 million of local currencies that AID hoped to spend to promote policy reform were left unspent because Tunisian offi- cials did not want to make some of AID’S proposed policy reforms to Tunisia’s agricultural credit fund at the time they were proposed. Page 21 GAO/NSIAD-BOSlOBR Foreign Assbtance Appendix I F9ogrammhg and Use of Local Currencies Tunisia is a lower middle income developing country, and the govern- ment places a high priority on structural adjustment (e.g. restructuring its economy to provide more opportunities for export-oriented growth). Tunisia’s adjustment program is aimed at reducing the deficit and gov- ernment spending, removing price controls, and liberalizing investment and foreign exchange controls. Tunisian officials are very concerned about high unemployment resulting, in part, from their adjustment pro- cess and the severe drought Tunisia experienced in 1987-89. AID mission officials believe that the Tunisian government is capable of and com- mitted to setting sound economic development priorities. According to AID and Tunisian officials, there is little disagreement on general devel- opment priorities. AID revised its assistance strategy for Tunisia in mid-1987. It no longer emphasizes traditional infrastructure and institutional development, but, instead, stresses economic reform in support of Tunisia’s structural adjustment. AID'S strategy is to provide Tunisia with balance-of-payment relief through EsF-funded commodity import programs and Public Law 480 title I concessional food credits, while using local currencies to pro- mote policy reform and to support government programs. In 1988, Tunisia received a total of $42.4 million in U.S. assistance, including $31.6 million through Public Law 480 titles I and II food aid, $10.8 million in ESF commodities, and a small amount ($28,000) in devel- opment assistance. Tunisia generated about $30 million in local currency in fiscal year 1988 from these programs, according to mission officials. The amount of local currencies generated and jointly programmed does not always equal the value of the assistance provided in any given year. It is not uncommon to generate local currencies in fiscal years following the receipt of the assistance, especially when commodities such as food aid are shipped or arrive in-country near the end of the fiscal year. In fact, in 1988, the U.S.-Tunisia Public Law 480 title I agreement was amended twice in the last half of fiscal year 1988. Page 22 GAO/NSIAD-90-210BR Foreign Assistance Appendix I ProgramtUg atlU Use of Local Currencies Plgunl.#z1@88LourlCurroncy tixpondltum InTunlrlr Forage Reserves ($4.0 million) 5.?Oh Export Promotion ($1.8 milliin) 81.5% l - Public Works/Employment ($25.7 million) Local currencies equivalent to $3 1.5 million were expended in Tunisia in fiscal year 1988, as shown in figure 1.9. Useof FoodAid43enerated LocA Local currencies spent from food aid-generated funds in fiscal year 1988 currency were used to support key Tunisian government rural public works projects. The projects were designed to directly increase the produc- tivity and income of the rural poor by providing temporary “safety net” employment for those adversely affected by Tunisia’s structural adjust- ment program and the severe drought. The projects supported US. assistance objectives by promoting agricultural and rural development activities essential to increased agricultural production, such as soil and water conservation, desert control, and reforestation projects. Local currency support for the public works projects fostered political goodwill between Tunisia and the United States. Tunisian officials value ongoing U.S. support for these projects because they view them as key to the success of structural adjustment. By providing employment, the officials believe the projects cushion the negative effects of adjustment and help prevent or forestall social unrest. In fiscal year 1988, local cur- rency generated from U.S. assistance funded more than 23 percent of the largest set of projects, which absorbed about 20 percent of Tunisia’s unemployment. Page 23 GAO/NSIAlMO-210BR Foreign Assistance APW* 1 prorpammine and Use of Local Currencies Useof ESF-Generated Local Local currencies generated from ESF commodity import programs were Currency used to support a Tunisian government agricultural development pro- gram, a government private sector program, and to promote policy reforms consistent with Tunisia’s adjustment program and U.S. open market private sector development objectives. Local currencies supported a government livestock forage reserve pro- gram. Tunisia’s livestock population dwindled as a result of the drought, and this program was designed to increase the availability of forage supplies for farmers affected by the current and future droughts. In late 1988, AID provided $1.8 million in local currencies to a Tunisian government export risk insurance company. In exchange, the govern- ment company agreed to increase its data base on non-traditional, but promising, markets in Latin America, Sub-Saharan Africa, and Asia. Also during 1988, the AID mission offered the equivalent of $6.0 million in local currencies to support Tunisia’s leading agricultural credit pro- gram, in exchange for four policy changes intended to improve the credit program’s financial status and to increase credit opportunities for poor farmers. Tunisian officials agreed to two of the proposed changes-to reduce subsidized lending to politically influential wheat farmers and to purge uncollected loans more than 3 years old. However, the officials maintained that the other two changes-reducing adminis- trative banking charges and establishing a more rigorous collections system-were not in the government’s best interest at the time. As a result, the mission released half of the requested funds, approximately $3.0 million in fiscal year 1989. The local currencies that were not released remained unused in a special account at the Central Bank, and had not been reprogrammed as of November 1989, according to AID offi- cials. Tunisian government officials said they were already considering these reforms independently, and probably would have made the same changes on their own, but would have waited a little longer to do so. EsF-generated local currencies were also to be used for an AID agricul- tural development project. In 1988, the AID mission and the Tunisian government established a project to analyze and formulate agricultural policy consistent with the objectives of Tunisia’s structural adjustment program. AID mission officials anticipated funding certain studies and projects with local currencies generated from commodity import pro- grams As of August 1990, no local currencies had been used for this project, according to an AID mission official. Page 24 GAO/NSIAD-8@21OBB Foreign A~Mance Appendix I Programmlng and Uee of Local Currender Flguro 1.10 w Zaire: Was Use Consistent With Assistance Objectives? l Generally consistent l Currencies were used for: mAIDdevelopment projects @AIDmission support *Zaire budget support l Program designed to cut fraud and abuse In addition, some local currencies from prior year agreements remain unprogrammed or unexpended. We were told by Central Bank officials that Tunisia’s monetary authorities do not undertake compensatory monetary actions when currencies such as these are withdrawn from the money supply, either by increasing available credit or printing addi- tional currencies. Therefore, unused local currency in Tunisia represents a real loss to the money supply. Page 26 GAO/NSIAIMO-21OBR Foreign Assistance , Appendix I Frogrammhg and Use of I.ocal Currendea Local Currency Use in In fiscal year 1988, local currencies generated from US. food aid and ESF Zaire programs to Zaire were used in support of U.S. assistance objectives. Spending went primarily to finance the local costs of AID development projects, AID mission costs, and general budget support of the Zaire gov- ernment. Although the currencies are used for AID projects, they are owned by the host country, and AID and Zaire officials jointly manage the currencies’ use. According to both U.S. and Zaire officials, this has helped to reduce fraud and abuse, and foster better accountability con- trols and procedures. Zaire has one of the lowest per capita income levels among African countries. Presently, Zaire faces a growing external debt burden, depen- dence on commodity exports with unstable prices, physical infrastruc- ture deterioration, and inadequate public investment. Zaire implemented an economic stabilization program in 1983. However, in 1986 and 1987, the government undertook excessive deficit spending, and solid gains from the stabilization program eroded. In 1987, the World Bank and the International Monetary Fund withheld further loans until the govern- ment re-established its stabilization program in early 1989, but balance- of-payments support was again withheld in 1990 due to excessive gov- ernment spending. The most recent stabilization program emphasizes reducing the budget deficit, restructuring and improving public sector management, improving the business climate, and improving the trans- portation network. AID assistance objectives include economic restruc- turing and policy dialogue, increasing agricultural productivity, and improving farm-to-market access, health and population planning. In 1988, Zaire received $42.2 million in US. assistance, $29 million from DFA, and $13.2 million from Public Law 480. Zaire generated an equivalent of $30.9 million in local currency from title I of Public Law 480 and ESF commodity import programs. As in Tunisia, some of these generations were from prior year commodity import programs, and some local currencies from fiscal year 1988 assistance would not be gen- erated until future fiscal years. Page 26 GAO/NSLAD-f)O-21OBR Foreign Amiatance Appendix I Programmhg and Use of Local Currencies Expendlturer in Zaire AID Mission Support ($3.4 million) of Zaire Budget Support ($2.2 million) Budget Support: Training Institutions and Other ($1.5 million) 70.6%- - AID Development Projects ($17.2 million) AID Development Projects Figure I. 11 shows how $24.3 million in local currencies was spent in Zaire in fiscal year 1988. More than 70 percent of the local currencies supported dollar-funded AID development projects. These were primarily agriculture and rural development projects aimed at increasing food production, raising national incomes, and improving nutrition. Other projects focused on child survival strategies, family planning, and acquired immunodeficiency syndrome (AIDS) prevention. Some local currencies were used to support selected development projects in trans- portation, infrastructure, local private voluntary organizations, and the Peace Corps. Specifically, local currencies in support of AID development projects were used to purchase locally available goods, such as fuel and vehicles, and for training and salaries. Also, currencies were used to supplement the low salaries of Zaire government workers assigned to AID projects. In addition, AID and Zaire officials have agreed to place local currencies in high yield term deposits when generations tempora- rily exceed expenditures. In 1988, $4.1 million in local currency was deposited in term accounts due to high generations from prior year programs. Page 27 GAO/NSLAD-90-21OBR Foreign Assistance Appendix I Programmbg and Uae of Local Currencies AID officials saw many benefits in using local currencies to support dollar-funded AID development projects. According to mission officials, restrictions on spending appropriated dollars do not apply to local cur- rencies; therefore, using local currencies provides more flexibility to respond to project needs. Also, local currencies can be used from one year to the next without reobligation, and can be invested in short-term deposits when there is a periodic surplus. However, the AID IG criticized the AID mission in Zaire for a lack of gui- dance on procurement practices when using local currencies. AID IG audi- tors noted that project officials did not always document that multiple sources of supply had been sought for commodity purchases, and recom- mended that the mission prepare guidance to establish standard pro- curement procedures. Mission officials agreed to prepare additional guidance, although they maintained that flexibility is a key advantage to local currency use. MissionSupport The next largest share of local currency spending (14 percent) was for mission operating expenses and program support for the various offices managing the AID development projects. AID projects and the mission itself depend, to some extent, on local currencies for operating expenses. Mission planning documents for fiscal years 1990 and 1991 indicated that a loss of local currencies without an increase in dollar resources for AID development projects would prevent the mission from managing its current project portfolio. BudgetSupport Local currencies were also used for the government of Zaire’s budget support. Zaire’s planning ministry received funds for general expenses and planning, as well as operating expenses for a special office, the Sec- retariat of Counterpart Funds. The AID mission does not monitor the use of local currencies for budget support beyond ensuring that they are deposited into the proper government account, according to mission offi- cials. The Secretariat, staffed with 30 people, was formed to manage local currency resources generated by nonproject assistance programs, funded through the United States and six other donors. In addition, the Ministries of Health and Agriculture received local currency support, as did a few training institutions and local private voluntary organizations. According to AID officials, the Zaire government has a history of corrup tion and a lack of control over the use of public funds. By using local currencies for AID projects, the officials said, they can closely monitor local currency use and ensure spending for development purposes. A Page 28 GAO/NSLALMO-210BR Foreign b&ance Appendix I Progmnmlng and Use of Local currender 6387 AID IG audit found that the Zaire mission had an effective local cur- rency audit system in place and, during our visit, AID officials were taking further steps to ensure that projects were maintaining adequate controls over local currencies. Both AID and Zaire officials stated that joint accounting and monitoring requirements help reduce fraud and abuse. To further improve accountability, AID officials said they work closely with Secretariat of Counterpart Funds officials to improve oversight capabilities. AID monitors Secretariat reports, works with officials to improve report reliability, and provides formal training to Zaire govern- ment officials. Local currencies were used to purchase computers for the Secretariat, and AID’S software system for local currency accounting was being translated into French for use by the Secretariat, according to AID officials. However, the volume of currencies managed by the Secretariat in 1988 was small (2.2 percent) relative to Zaire’s overall expenditures. As a result, although local currency monitoring requirements may have contributed to fiscal responsibility within the Secretariat, they have not necessarily had an effect on the overall government of Zaire’s financial management capabilities or systems. Page 29 GAO/NSIAD@O-21OBB Foreign AmMance Appendix I Programming and Use of Local Currencies Figure 1.12 m Accountability Requirements l AID management and ‘IG differ on accountability l Local currencies are vulnerable to misuse l Focus should be on host country financial management capabilities must ensure that local currencies are used for the purposes outlined Local Currency AID by the Congress and in the aid agreements. However AID and its IG differ Accountability on the degree of accountability and control AID missions should exercise Requirements over local currencies. Our previous reviews and AID IG audits have found that local currencies are vulnerable to misuse in many countries, and AID is considering additional oversight requirements to strengthen assur- ances that local currencies are properly used. Legislation requires that most host country-owned local currencies be used for development purposes. AID guidance provides missions with flexibility in determining the degree of oversight necessary to provide Page 30 GAO/NSIAD-BO-210BR Foreign Assistance Appendix I Progmnndng and Use of Local Currencies reasonable assurance that the local currencies are used for the intended purposes. The guidance permits missions to vary their oversight, depending on their involvement in programming the local currencies and their assessment of the management capabilities of the host country. AID management and the AID IG differ on the extent of accountability and control AID missions should exercise over these local currencies. AID offi- cials stated that the disagreement over the necessary level of accounta- bility has created frustration and uncertainty at AID missions. AID argues that, although it must be satisfied that local currencies are used for appropriate purposes, host countries, and not AID, ultimately should be accountable for the proper use of the currencies because they own them. AID officials maintain that increasing demands for accountability create friction with the host government because government officials view the currencies as their own. For example, language inserted in Public Law 480 title I agreements in recent years permits U.S. officials to audit host country records. We found that Tunisian officials strongly objected to this provision, which they viewed as obtrusive and unnecessary. Also, some AID officials have told us that they simply do not have enough staff to significantly increase their local currency monitoring. AID'S IG believes that AID missions must maintain full financial accounta- bility for the local currencies, because they are generated from U.S. assistance. Based on this criterion, IG audits have found that many AID missions, because of accounting and monitoring weaknesses in the mis- sion or in the host government, cannot always assure that local curren- cies are properly used. Our previous reviews also found that because of inadequate accounting, monitoring, and reporting systems, missions could not always determine whether withdrawals and disbursements were made for agreed purposes. Also, we found in a recent review of assistance to El Salvador that mission officials and IG auditors do not always interpret AID guidelines similarly. AID officials acknowledge that improvements could be made in some areas, while still holding the host country ultimately accountable. A 1989 AID survey of mission procedures identified several key problems. AID officials told us that they are considering issuing additional over- sight guidance intended to address these problems and decrease the vul- nerability of local currency to misuse. The changes include strengthening host country reporting requirements and mission verifica- tion procedures for local currencies held in special accounts, requiring Page 31 GAO/NSlAD-90-210BR Foreign Assistance Appendix I Progmnunhg and Use of Local Currencies audits of host country agencies managing the accounts and the organiza- tions receiving the funds, and requiring formal and standardized finan- cial assessments of host country agencies managing the accounts and those receiving the funds, when necessary. On April 18, 1990, we testified before the Subcommittee on International Economic Policy and Trade, House Committee on Foreign Affairs, in support of these changes. The audits and the formal financial assess- ments could be funded with local currencies, which would result in improved accountability without further taxing scarce mission resources. The proposed changes would help to place the focus for local currency accountability oversight on assessing and improving host country agency financial management systems overall, rather than on individual misuses and deficiencies. Page 32 GAO/NSIAD-90-21OBR Fore&n Assistance Appendix I Frogranunhg and Use of Local Currencies Figure 1.13 GM Local Currency Factors Affect Host Country Economies Local currencies can impact on host country economies, depending on: @Timebetween generation and expenditure l Local currency volume relative to money supply @Hostgovernment monetary policy Generating and spending local currencies can have a macroeconomic Did Local Currency effect on the host country economy, depending on (1) the time elapsed Use Impact on Host between the generation and expenditure of local currencies, (2) the size Country Economies? of local currency generations and expenditures compared to certain macroeconomic variables such as money supply, and (3) the monetary actions taken by the host government to compensate for potentially adverse effects. When a recipient country sells U.S.-provided commodities (normally to the private sector) in exchange for local currencies, the currencies are Page 33 GAO/NSIAD-z)o-21OBR Foreign Assistance , Appendix I Programmhg and Use of Local Currencies withdrawn from the private sector and deposited in a government account, usually in the central bank.’ As a result, the money supply con- tracts, resulting in a deflationary effect. On the other hand, when these local currencies are spent, the money supply expands, which may result in an inflationary effect. The potential macroeconomic effect of this local currency generation and spending can be offset if spending closely approximates genera- tions. If the generation and expenditure of local currencies in these countries occur at about the same time, then the contraction and expan- sion effects may cancel out any net macroeconomic effect. Am’s overall guidance to missions states that local currencies should be disbursed as quickly as is consistent with sound programming and prevailing eco- nomic conditions in the recipient country. Currencies are not always timely spent and, in some cases, they accumu- late over a period of years. The volume of local currency generated and spent influences whether this activity is likely to have a macroeconomic effect. When the volume is small, relative to certain macroeconomic variables such as a country’s broad money supply,2 the macroeconomic effect may not be discernible. However, the larger the volume, the greater the potential that this activity can have a significant macroeconomic effect in the host country. These macroeconomic effects may not always have an adverse effect on the host country economy. For example, generating, but not spending, local currency may reduce any existing inflationary pressures. However, if these accumulated cur- rencies, sometimes referred to as the “pipeline,” are then spent over a short period of time, they can have an inflationary effect, depending on the volume of the accumulation. Finally, host country monetary policy can affect the impact of local cur- rency activity. When commodities are sold and the money supply is reduced, government monetary authorities can take compensatory action to offset this reduction by increasing available credit, or even printing more currency. On the other hand, when the local currencies are spent and the money supply expands, monetary authorities can take compensatory action by restricting available credit. The monetary ’ In Zaire, local currencies generated from U.S. assistance are deposited into a government account at a commercial bank. 2Broad money supply primarily includes currency and checking and savings accounts owned by the general public. There are numerous macroeconomic variables that could be used for this comparison. We chose broad money supply because it is a commonly used indicator of an economy’s financial size. Page 34 GAO/NSIAD-BOflOBR Foreign Assistance Appendix I Programming and Use of Local Currencies authorities of the individual country will determine whether, and at what point, these actions are taken. Figure 1.14 0 Did Local Currency Use impact on Host Country Economies? l Overall macroeconomic impact unlikely in 1988 because: 4olume was relatively small Gurrencies were generally spent soon after they were generated lPotential exists in some countries lOther donor currencies can be a factor @AIDguidance is minimal Unlikely Macroeconomic Impact Using data on the 1988 volume and the timing of currency expenditures From U.S.-Generated Currencies for 47 countries, we assessed the potential for local currencies generated from U.S. assistance to have a macroeconomic impact on host country economies. First, we examined whether local currency activity was large enough to have a potential macroeconomic effect by calculating, for 4 1988, the amount of local currency generated, spent, and the resulting year-end accumulations as a percent of host country broad money supply for the 47 countries where data was available. As illustrated in table I. 1, on average, the generation, spending, and resulting pipeline Page 36 GAO/NSIAD-!bOH)ISOBR Foreign Assistance . Appendix I procprunmine and Use of Local Currencies were near or less than 2 percent of the broad money supply. There is no generally accepted threshold beyond which the volume of local currency generations and expenditures is large enough to have a significant effect on a country’s economy. However, it is unlikely that local currency gen- erations and expenditures of this size in and of themselves would have a significant macroeconomic effect in these 47 countries. Table 1.1: 1988 Local Curronoy QOnOratlOn, Expondlturo, and PipOlIn I# Figures in percent a Percent of Broad Money Supply Pipeline Generation Expendlture year end Average for 47 countries 1.81 1.77 2.02 Countries with significant Costa Rica 9.0 El Salvador 10.4 Costa Rica 12.0 ratios Bolivia 7.8 Costa Rica 9.1 Lesotho 6.9 El Salvador 6.8 Note: Individual countries are listed if the percent of local currency activity compared to the broad money supply is more than two standard deviations from the mean. Table I. 1 shows that local currency generation ratios in Costa Rica, Bolivia, and El Salvador; spending ratios in Costa Rica and El Salvador; and the pipeline ratios in Costa Rica and Lesotho were substantially larger than the average. We do not know whether these volumes are large enough to have had a discernible macroeconomic effect in these countries because we did not do an in-depth analysis in each country. But, the large local currency ratios identify countries where the poten- tial may exist for local currency generations and expenditures to have a macroeconomic impact. For example, if the pipeline in Costa Rica were spent over a short period of time, the volume of accumulation is large enough to suggest that this expenditure could have a discernible infla- tionary effect on the country’s economy. We also calculated for these 47 countries whether local currencies were spent at about the same time that they were generated, using 1988 expenditures minus generations.3 If currencies are spent soon after they are generated, the macroeconomic effect can be negated. When calcu- lated as a percent of the broad money supply, the data in table I.2 shows that on average, these countries spent a mere 0.06 percent more than they generated. With this, we concluded that, because most local curren- cies generated in 1988 were also spent in 1988, missions were in compli- ance with AID guidance and possible macroeconomic effects were cancelled out. Therefore, in most cases, generating and spending local 3Annual figures for expenditures less generations were the only readily available data. We do not know if this is the optimal period of time to consider. Page 36 GAO/NSIAD-30-21OBR Foreign Assietance Appemdix I Programming and Use of Local Currencies currency from U.S. assistance most likely did not have a macroeconomic effect. Table 1.2: 1988 Net Local Currency Expenditure a8 a Percent of Broad Figures in percent Money Supply Expenditure Less Oeneration Average for 47 countries -0.04 Countries with significant ratios El Salvador 3.6 Madagascar -3.6 Bolivia -3.7 Malawi -5.1 Note: Individual countries are listed if the percent of net local currency expenditure compared to the broad money supply is more than two standard deviations from the mean. For example, as shown previously in table I. 1, in Costa Rica the timing of 1988 generations and spending offset possible macroeconomic effects. Even though the relative size of generations (9 percent) and expendi- tures (9.1 percent) were substantially larger than the average, expendi- tures exceeded generations by only 0.1 percent. By spending the local currencies in 1988 close to the time they were generated, the potential macroeconomic effect has been minimized. However, in El Salvador, spending outpaced generations, and the relatively large ratio suggests that this may be an example of where spending could have had an infla- tionary effect. Finally, as shown in table 1.2, in Malawi, Madagascar, and Bolivia, the volume of local currencies generated and not spent, compared to the broad money supply, deviated significantly from the average. These unspent currencies may have reduced inflationary pressures in these 1 countries, However, the resulting currency accumulations could con- tribute to inflation at a later date if care is not taken in planning and monitoring future use. We did not conduct an in-depth analysis of mone- tary policy or other macroeconomic factors in these countries, and do not know whether local currency generation and expenditure actually did have a macroeconomic impact. MacroeconomicEffect in Tunisia We analyzed (1) the volume of local currencies relative to the money andZaireUnlikely supply, (2) the time between generations and expenditures, and (3) host government monetary policy in Tunisia and Zaire. We believe that AID'S programming and expenditure of host country-owned local currencies, in fiscal year 1988, most likely did not have a significant macroeconomic effect. Page 37 GAO/NSIAIMO-210BR Foreign Amistance Appendlx I Frogranuning and Use of Local Currencies In Tunisia, the ratios of local currency generation, expenditure, pipeline, and timing of expenditure compared to broad money supply, were all under one percent and it is unlikely that local currency generation and expenditures would have had a discernible macroeconomic impact. In Zaire, we found that the ratio of local currency expenditures to money supply was almost 6 percent. However, the mission in Zaire, working with government officials, instituted a planning and monitoring system, which ensured that the local currencies were spent soon after they were generated, thereby mitigating possible macroeconomic effects and the need for compensatory monetary actions. Potential Exists for a Although in most cases, U.S.-generated local currency amounts were Macroeconomic Impact small in 1988 and currencies were spent soon after they were generated, local currency activity could have a macroeconomic effect in some coun- tries. We identified countries where the potential existed for local cur- rency generations and expenditures to have had a macroeconomic effect. However, we did not perform in-depth analyses in these coun- tries. Also, the combined generations and expenditures of local curren- cies from other donor assistance can be large and could have a significant macroeconomic effect. For example, in Zaire in 1989, local currencies generated from U.S. assistance comprised less than 40 per- cent of total local currencies generated from bilateral donors, according to a planning ministry official. This does not include local currencies generated through World Bank assistance. Although our analysis showed no likely macroeconomic effects when considering U.S.-gener- ated local currencies alone, we did not determine the combined effect considering other donor generations and spending. AID recognizes that local currency generation and spending can have a macroeconomic effect. AID guidance calls for spending local currency as quickly as possible, given conditions in the host country and considering stabilization agreements and sound monetary policy. However, the gui- dance does not provide further detail on what factors to consider in ana- lyzing host country conditions and monetary policy, or on appropriate mission responses in those cases where it is not possible to spend local currencies quickly. Also, guidance is silent on those examples where local currency accumulations could help reduce inflationary pressures. AID'S Africa Bureau has issued guidance calling for a more complete examination of the macroeconomic implications of local currency activity and for programming in the context of government monetary policy. Page 38 GAO/NSIAD-BO-21OBR Foreign Assistance Appendix II Comments From the Agency for International Development AGENCY FOR INTERNATIONAL DEVELOPMENT WASWINOTON. D c 20123 ASSISTANT AOMINl5TRATOR 4JG 3 iclfm Mr. Frank C. Conahan Assistant Comptroller General National Security and International Affairs Division General Accounting Office Washington, D. C. 20548 Dear Mr. Conahan: The Agency for International Development (A.I.D.) concurs with the findings and conclusions in the GAO draft report, “Foreign Assistance: Use of Host Country-owned Local Currencies.” I wollld like to highlight the four key conclusions of the report. 1. When the United States provides commodities or dollars as assistance to recipient countries, the local currencies that are generated from this assistance are owned by the host country. The currencies are not an added resource to the host country economy -- the added resource comes with the initial U.S. assistance provided in the form of cash or Now on p. 2 commodities. (p. 3) This conclusion is consistent with the existing statutory framework of the Foreign Assistance Act of 1961, legislative history accompanying appropriations legislation, and years of Agency practice recognized by Congress. 2. In Tunisia and Zaire [the two countries in which the GAO undertook fieldwork], local currencies were programmed in fiscal year 1988 for uses generally consistent with U.S. development objectives and were generally used for the Now on p. 2 agreed upon projects and programs. (p. 4) I believe this conclusion also applies to most countries in which local currency is programmed. 3. A.I.D.‘s oversight of local currency use has not always been adequate to reasonably ensure that the currencies were consistently used for the agreed upon purposes. A.I.D. is considering improved oversight guidance that will properly focus on the responsibilities and financial management Now on p, 2 capabilities of the host government. (p. 4) I am pleased Y J Page 39 GAO/NSIALk90-210BR Foreign Assistance , Appendix II Comments From the Agency for Inti3nationd Development -2- to report that A.I.D. ‘6 Office of Financial Management has nearly completed the preparation of stronger, clearer Agency guidance that will improve local currency oversight. 4. In most countries, it is unlikely that in 1988 generating and spending local currencies associated with U.S. assistance had a significant macroeconomic effect. Also, in most countries accumulations of unspent local currencies are not large enough to have a significant future Now on p. 2. macroeconomic effect. (P. 4) The draft report represents a comprehensive and balanced assessment of the use of host country-owned local currencies. The analysis is professional and the presentation of the results, especially the use of graphics, makes the document a readable and valuable resource for future reference. I appreciate the opportunity to comment on the report. Sincerely, BUteaU fOK PKOgKam and Policy Coordination cc: IG/PPO, John Eckman Page 40 Apperidix III Cixnments From the Department of Agricdture supplementingthoseinthe report text appear at the end of this appendix. UnltedStatea Foreign Washlngton,D.C. ;f$upral 20250 Mr. Frank C. Conahan Assistant Comptroller General International Affairs Division U.S. General Accounting Office 441 G Street, N.W. Washington, D.C. 20548 Dear Mr. Conahan: This Is In response to your recent memorandum requesting our revlew of draft report NSIAD-90-210BR entltled "Foreign Assistance: Use of Host Country Owned Local Currencies.“ We concur In your overall conclusion that AID Is responsible for providing assurance that host government owned local currencies be programned for agreed upon purposes. The guidelines proposed by AID's financial management offfce in April 1990 to address the weaknesses pointed out in this draft report, which include a requirement that host countries using local currencies submit formal and standard financial assessments and audits of host country agencies, should improve AID's oversight of host country programs. More speclfically, the Department offers the following comments regarding the report: Seecommentl. 1. References to Title I, P.L. 480 "loans" should be changed to "concessional credit" wherever they appear. Seecommentl. 2. References to sectlon 416 should be changed to "416(b)". Nowonp.14. 3. The second paragraph on page 22 should be revised to indicate that trust accounts are uncommon in Title I and section 416(b) programs. This paragraph glves the impression that trust accounts for local currencies are common. There Is only one P.L. 480 case of which we are aware where a trust Seecommen;2. account was used. In this case, the country did not have the sophistication to manage the currencies generated. There may be some confusion between trusts and special accounts. Finally, one area of your report which we would like to address concerns the programming of host country owned local currencies for AID mission general operating expenses and costs associated with supporting development projects. Y Page 41 GAO/NSL4D-9O-21OBR ForeQn Adstance , r Appendix III Comments From the Department of Agriculture Mr. Frank C. Conahan 2 Uhlle we appreciate that using these currencies in this manner may be approprlate in certain instances, we belleve the general Intent of the See comment 2. P.L. 480 legislation Is to provide direct support for agreed upon development programs and projects. Thank you for the opportunity to comment on this draft. Slncerely, Administrator Y Page 42 GAO/NSIAIMO-210BR Foreign Assistance Appendix KU comments From the Department of A@lculture The following are GAO’S comments on the Department of Agriculture’s letter dated August 3, 1990. 1. The wording in the text has been revised to incorporate this suggested GAO Comments change. 2. Local currencies held in trust by the missions and used for mission administrative costs are primarily generated through ESF or DFA. Curren- cies generated through Public Law 480 must be used for development purposes, while a portion of currencies generated through ESF and DFA ’ may be used for other purposes such as administrative expenses. ’ Page 43 GAO/NSLAD-BO-21OBR Foreilpl AssMance C , Appendix IV Mqjor Contributors to This Briefing Report ’ Donald Patton, Assistant Director Nationa1 Security and Kay E Brown Senior Evaluator International Affairs ’ Bruce L. Kutnick, Senior Economist Toni Y. Townes, Evaluator Division, Maria Santos, Evaluator Washington, D.C. Page 44 GAO/NSIAD&21OBR Foreign Assistance Page 46 GAO/NtRAD@O-210BR Foreign Aseistance Page 40 GAO/NSLAD-SOdlOBR Foreign Assistance Page 47 GAO/NSIAD-BO-210BR Foreign Aseletance Related GAO Products El Salvador: Accountability for U.S. Military and Economic Aid (GAO/ NSIAD-90-132, Sept. 1990) Using Local Currency Generated by U.S. Food Aid for Development Pur- poses (GAO/T-NSAID-90-32, April 18, 1990). Foreign Aid: Problems and Issues Affecting Assistance (GAO/NSIAD-8% 61~R, Dec. 30,1988). Foreign Aid: Better Management of Commodity Import Programs Could Improve Development Impact (GAO~NSIAD-88-209, Sept. 26,1988). Food Aid: Improving Economic and Market Development Impact in African Countries (GAOINSIAD-88-65, Dec. 21, 1987). Foreign Aid: Accountability and Control Over U.S. Assistance to Indo- nesia (GAOjNSIAD-87-187, Aug. 19, 1987). Liberia: Need to Improve Accountability and Control Over U.S. Assis- tance (GAO~NSIAD-87-173, July 16, 1987). (472195) Page 48 GAO/NSLAD-90-210BR Foreign As&tame .--. I. .I.._..__...._ - ._... - .._.. -._--- .-- .._.._ .._.._ .__ - ..__-_____ -_ ..___I__.__ ___~__ - -_.- -_-__-l.---*_l”l~-~~.-~~~-~ --.a *.. Orclthrs may also bv placd by calling (202) 275-6241. I._-- .l_-.--- -...-_- ..-.---~-.- First4 :Iass Mail I”ost:kJLFt~XLI’cv~s Paid (;A() l’c~rrruit No. (; 100
Foreign Assistance: Use of Host Country-Owned Local Currencies
Published by the Government Accountability Office on 1990-09-25.
Below is a raw (and likely hideous) rendition of the original report. (PDF)