Publications

	Table of Contents
p. 3 of 4 BACK | NEXT I. Introduction II. Background III. Proposed Approach IV. Conclusions and Recommendations

Review of Lending Foreign Exchange for Local Currency Expenditures on Projects

---

III. Proposed Approach

A. Need for LCF

33. An essential function performed by the Bank as a development financing institution is to transfer financial resources from capital-surplus countries to the Bank's DMCs that are experiencing resource constraints. The Bank's LCF, particularly under ADF, has been largely a vehicle for relaxing the resource constraints facing the Bank's poorer DMCs. However some LCF, particularly to the PRC from OCR, has bee made on the basis of the Bank providing a reasonable share of project costs.

34. Resource constraints for most of the DMCs are expected to continue through the 1990s. While it will be important for the Bank to continue to play its crucial role in international financial intermediation, as a development financing institution it should also ensure that the borrowers' commitment and ownership are adequately reflected in Bank financing policies. Moreover, the Bank wishes to playa more proactive role in projects that address cross-cutting concerns in social development. These projects usually have a substantial local cost content. The Bank's LCF policy in the future should keep these considerations in view.

B. Comments by the Board

35. Recent Board discussions on several loan proposals involving LCF have included comments on the Bank's LCF policies and practices. For example, a request was made for a historical analysis of LCF by country and by sector. The high proportion of LCF as a percentage of the total local cost component of the project has been questioned on several occasions. Questions have also been raised concerning the comparability of the Bank's policies with those of the World Bank regarding indirect foreign exchange and LCF. It was commented that the Bank tended to start automatically with the standard percentage limit and include a large proportion of LCF. Views have also been expressed that DMCs with a favorable balance-of- payments situation and large foreign exchange reserves do not merit LCF. The Board has also been stressing that LCF should not substitute for domestic resource mobilization by the borrower, and that LCF policy should discriminate in favor of projects in the social sector. Some Board members have proposed that, in calculating project costs for determining the standard percentages of financing, the Bank should exclude taxes and duties, as is the practice of the World Bank. A few Board members have also questioned the validity of distinguishing between local currency expenditures and foreign currency expenditures in the context of liberalizing exchange regimes in DMCs.

36. This Paper has presented a historical analysis of Bank LCF, showing that, on a l post facto basis, the Bank's LCF has been consistent with its policy intentions and within the i Charter's limitation. In so far as the high percentage of Bank financing of the local cost component of some projects is concerned, some project-specific observations may be helpful. In several cases, the size of the Bank loan w determined in relation to an appropriate level of project financing rather than to the foreign exchange component, which was low. For example, the $200 million Bank loan made in 1992 for the Guang-Mei-Shan Railway¹⁶ Project in the PRC included $82 million in LCF. However, the to at Bank loan represented only 37.5 per cent of the total cost of the project, which provided infrastructure required for the growth and development of the mountainous and less-developed eastern area of Guangdong Province. The executing agency of the project, Guangdong Guang-Mei-Shan Railway Company, financed $333 million equivalent through a combination of domestic borrowing, bond issues, and internal funds. Therefore, the LCF element in the Bank and was based on cost-sharing considerations. Similarly, while the Bank provided $7.3 million for LCF in its $10.4 million loan for the Tourism Infrastructure Development¹⁷ Project to Nepal in 1992, the Bank loan amounted to a little more than 70 per cent of the total project cost. Similarly, while the $20.0 million Bank loan to Sri Lanka in 1991 for the Low-lncome Housing Development¹⁸ Project included $9.5 million in LCF the loan represented only 60 per cent of the total project cost.

37. As regards the balance-of-payments implications of Bank-assisted projects, whatever portion of project cost that is not finance by the Bank will probably result in diversion of resources to the project from the other sectors of the economy and a widening of the resource gap, There are several policy options for reducing the gap, but there will be a lag between the policy action and the desired impact on the resources situation. In the short run, the resource situation will likely worsen. Where the concerned DMC faces a balance-of-payments constraint, financing a higher proportion of the project cost than the foreign exchange component will provide balance-of-payments relief, A favorable balance-of-payments situation or accumulation I of foreign exchange reserves over a short period does not necessarily indicate a long-term, f sustainable equilibrium in the balance of payments, Moreover, external sector adjustment should r be sustainable in both current and capital a counts of the balance of payments. The Bank's approach is that, as the DMC makes progress in external sector adjustment, the need for LCF will be reviewed and the provision of LCF will be appropriately adjusted, This approach seems appropriate for the future also.

38. As regards the continuing need for a LCF policy in an environment of liberalizing exchange regimes in DMCs, the distinction between local currency and foreign currency has weakened in many DMCs as consequence of recent economic reforms and greater openness, including trade liberalization, enlarging private sector role, and direct foreign investments. The Bank has supported and will continue to support the further liberalization of foreign exchange regimes in the DMCs. However, even though most DMCs have introduced some degree of exchange regime flexibility for current account transactions, exchange controls are prevalent to some degree. While there has been a definite shift towards integration of DMC economies with the global economy, the reform process is not yet complete. Until the foreign exchange controls are completely withdrawn all borrowing DMCs and their currencies are made fully convertible, the distinction between domestic currency and foreign currency will continue to be relevant in the management of these economies. A related question is whether the LCF policy would be relevant if cost-sharing guidelines were prescribed at the country and project level. The Bank's Charter requires that Bank resources be selectively used for financing local costs. In view of the prevailing exchange regimes in DMCs, it continues to be relevant for the Bank to distinguish between local currency expenditures and foreign costs. The Bank, by selectively limiting the financing of local costs' encourages domestic resource mobilization, fosters ownership and commitment by borrowers for development projects, and gives impetus to social development in its DMCs by preferentially supporting certain sectors and projects.¹⁹ Given the Charter limitation, the Bank's LCF policy is therefore a useful instrument for achieving the Bank's development objectives in its DMCs.

C. Indirect Foreign Costs and LCF: A Comparison with World Bank Practice

39. The Bank's policy on indirect foreign cost is similar to that of the World Bank. The calculation of the foreign exchange costs of a project in both institutions involves the following: (i) direct procurement of foreign goods and services; (ii) direct use of foreign resources by foreign contractors; (iii) direct use of foreign resources by local contractors; and (iv) indirect use of foreign resources by foreign and local contra ors, suppliers, and consultants. In turn, the last item includes (i) locally procured goods of foreign manufacture (use cost, insurance, and freight [CIF] value); (ii) estimated CIF value of major imported items (raw materials, components, etc,) used in the manufacture of domestically produced goods (this may include an estimate of the depreciation of costs of imported capital assets used in local manufacturing); (iii) depreciation of imported capital items used plus any foreign exchange payments to expatriate personnel under domestic consulting services and experts; and (iv) estimated imported materials of locally made components. As a practical matter, only items of significant size are normally included, and cost items that will have an inconsequentiaI effect on the final estimate and are difficult to estimate are disregarded.

40. As occurs in the World Bank, the Bank does not distinguish between direct and indirect foreign exchange costs in its financing. With the increasing openness to the world economy and the active participation by foreign investors and business enterprises in DMCs, the need to make such a distinction has become weak at best.

41. The LCF policies of the Bank d the World Bank are largely similar, which is understandable. The cost-sharing ceilings use by the World Bank (see Appendix 6) appear higher than the Bank's standard percentage limit for similar DMCs. However, the two ratios are computed differently. For example, the World Bank excludes duties and taxes in computing its ratio, while the Bank does not. Thus, a comparison is not strictly valid. Nevertheless, in actuality, the two ratios may represent similar values.

42. The main rationale for including taxes and duties in the cost estimates while i calculating standard percentages for Bank financing is that in most cases the payment of taxes and duties is a financial cost to the project, even though at the national level taxes and duties are a transfer payment and are not an “economic” cost to the project. Allocation of funds for payment of taxes and duties for a project shows commitment at the project level to give priority to that project in resource allocation. This is particularly true where the borrower and the project owner are not the same. For example, customs duties levied by the national government are a cost to a project implemented by the provincial government or an autonomous entity .The provincial government or that entity has the option either to import the concerned equipment on payment of customs duties or not to import the equipment and allocate these resources to some other use. Since the Bank is mainly a project-financing institution, commitment shown at the project level through allocation of taxes and duties is an important consideration for the Bank in deciding to assist the project. Accordingly, the present practice of including taxes and duties in the cost estimates while calculating standard percentage limits for Bank financing is justified. It is however recognized that in some cases he member-borrower and the project owner may be the same. In these cases, payment of taxes and duties may not represent a financial burden at the project owner's level as in the example discussed earlier. It will therefore be appropriate to determine the actual cost-sharing limit at the project level, keeping in view the relationship between the project owner and the borrower and the need to obtain an adequate level of commitment to the project from the Bank's counterparts, at the project owner's level or the borrower's level, as may be relevant.

43. The two institutions have similar practices regarding the minimum contribution required by the borrower to demonstrate commitment to a project. For individual projects, both banks usually expect the borrower to make a minimum contribution to project costs.²⁰ The World Bank defines this as net of taxes and duties, while the Bank does not specifically do so.²¹ However, both institutions find a lower contribution justifiable if government finances are seriously strained, and if warranted by a reassessment of the lending priorities. In general, to ensure that there is no disincentive to seeking cofinancing and to reduce the potential for conflict with other agencies, neither institution prescribes an external cost ceiling, allowing a government also to i demonstrate commitment by allocating to the project other external resources for cofinancing. Recent studies on the issue of "local ownership” and the quality of projects have pointed to the need to pay attention to a number of factors and issues to ensure commitment. However, while direct financial contribution cannot be treated as the all-encompassing proxy for commitment, it continues to be important that a minimum level of local contribution is secured to promote ownership.

44. In one respect, the World Ba k's LCF policy seems less restrictive than the Bank's. The World Bank's cost-sharing limits re applicable not to individual projects but to its overall lending program for a country, excluding financial intermediary, supplemental, emergency recovery, adjustment, and technical assistance loans. To ensure flexibility, the country limit is applied to a rolling three-year investment lending program, encompassing two prior years and the current year, However, the limits are ceilings, and actual cost sharing is expected to be below the limits. Because of the World Bank's ceilings, and since the Bank allows its financing to exceed the standard percentage limit in special cases, in reality this difference may be less relevant.

Back II. Background	Next IV. Conclusions and Recommendations