Journal Issue
Share
Article

The concessionality of foreign assistance: A new perspective on measurement of this concept

Author(s):
International Monetary Fund. External Relations Dept.
Published Date:
March 1984
Share
  • ShareShare
Show Summary Details

Toward a better understanding of this concept

Danny M. Leipziger

The basis for foreign assistance lending is that wealthier countries are willing to transfer resources to poorer nations at some cost to themselves. It is assumed that this cost—namely, the opportunity cost of capital—will be outweighed by the benefits the capital confers on the receiving country. Only if capital is transferred from a lower-return activity to a higher-return endeavor can global welfare be improved. This distinction between the rates of return on capital in the aid-giving country and the aid-receiving country is critical to a full understanding of concessionality in lending. In particular, it serves as the basis for ascertaining the circumstances under which a greater volume of concessional lending is preferable to a smaller amount of pure grant aid. It is also central to devising ways of extending the value of foreign aid.

To understand the definition of concessionality, consider that a $100 loan for ten years at 4 percent interest is “worth” almost $35 to a borrower who would otherwise pay 15 percent a year to borrow from the market. Since the net present value of repayments by the borrower totals $65, the grant element benefit is $35; to the borrower, the $100 loan is equivalent to a $35 grant and a $65 loan. But the lender faces a different opportunity cost of capital—say, 10 percent—and he receives repayments worth approximately $78 in light of his valuation of capital. Thus, the grant he is offering is $22, or 22 percent of the loan. In this case, therefore, the loan makes economic sense. It transfers resources to a more productive environment for capital, yielding a net benefit to society of $13. It should also be noted that the borrower is essentially indifferent between a grant of $35 and the aforementioned loan; however, the lender much prefers to offer the loan, since the effective cost to him of its $35 grant element is only $22.

Historically, this distinction between the benefit and cost of the grant element of a concessional loan has been blurred. Clarification could well lead to terms of assistance that are of greater value to recipients, while imposing no greater cost on benefactors.

Calculating grant elements

The primary factors that determine the “grant element cost” to lenders and the “grant element benefit” to borrowers are the final maturity, the grace period, and the interest charges of the loan, as well as the discount rate (opportunity cost of capital) to either the lender or the borrower. For simplicity, consider a conventional loan—immediately disbursed and repaid in equal installments—although these assumptions can easily be relaxed. Chart 1 shows the effects on the grant element of changes in these three basic parameters—maturity, grace, and interest charge—for a donor whose opportunity cost is 10 percent and a borrower whose rate of return on capital is 15 percent.

Chart 1Grant element calculations

A number of insights emerge. The first is that various combinations of terms can yield the same grant element level. This has important ramifications for foreign assistance policy. Second, it is clear that within a plausible range of variation, the interest charge affects the grant element calculation most significantly. Thus, for example, donors wishing to increase concessionality could do so most effectively by changing interest charges rather than maturities or grace periods. Third, it is clear that grant element calculations are extremely sensitive to the discount rate used. Conventionally, donors belonging to the OECD’s Development Assistance Committee have used 10 percent as their opportunity cost of capital in gauging concessionality. But this discount rate far exceeded the average weighted cost of capital of major donors in the 1960s and, conversely, underestimated it in the 1980s.

Another complication in the calculation of grant elements emerges once the assumption of immediate disbursement is relaxed. In fact, loan disbursement profiles vary considerably. A structural adjustment loan by the World Bank, for instance, is disbursed very quickly and would yield a larger grant element benefit to a borrower than a project loan that disburses in accordance with the gestation period of the project. The difference can be quite substantial: a standard IDA credit for $100 made to a borrower whose return on capital is 15 percent would entail a grant benefit of $91 if immediately disbursed, but only $42 if disbursed according to the average IDA disbursement profile.

Disbursement patterns have a similar effect on the calculation of the grant element cost, depending on the use to which donors put the eventually disbursed funds. Therefore, the concessionality actually offered by donors for project financing may be up to half what it would be for nonproject loans that are immediately disbursed. In this sense, development institutions are in a peculiar situation because they often receive their funding from donors as it is needed for actual disbursements. Since the institution itself does not benefit from lagged disbursements, it would be inappropriate to recalculate the grant element on the part of the lender qua disbursing agent. Nevertheless, there is a clear over-estimation in the conventional grant element calculations that assume immediate disbursement; these could be largely corrected if donors were required to deposit all contributions to international development institutions at the time of loan signing and if loan agreements were restructured to pass the funds on more rapidly.

Measuring the degree of concessionality of lending is therefore more complex than is generally acknowledged. In addition to distinguishing between the grant equivalent costs and benefits, more precision is required in the assumptions about disbursements, and, even more important, about the discount rate being used by both donors and recipients. As a practical matter, this will raise problems with respect to the appropriate value placed on capital by the borrower. Rates of return will usually not be reflected by the local cost of capital; capital markets in developing countries are rarely developed to the extent that capital costs reflect its scarcity value; moreover, interest rates are also frequently administered by the authorities. Still, the basic premises surrounding the supply and demand of concessional lending are well worth investigating.

Theory of grant elements

It is clear that there are potential advantages for both donors and recipients in exploring variations in terms, for any number of reasons. Consider, for example, the situation of developing countries facing liquidity problems. For cash-flow reasons, which in an ideal conceptual framework would affect the borrower’s discount rate, a borrower may prefer a longer grace period on a new loan and would be willing to accept a shorter final maturity or a higher interest charge. Similarly, on the part of donors there may be binding constraints on the volume of assistance that can be offered, but greater flexibility on the terms.

Based on economic theory, borrowers generally wish to maximize the grant element benefit of concessional loans, while lenders attempt to limit their grant element cost. Each calculation of grant elements is based on the terms of the loan and the relevant discount rate. The simple illustration in Chart 2 shows which combinations of interest charges and final maturity will yield the same grant element cost when the grace period is ignored and the discount rate is assumed to be 10 percent. The lender’s trade-off between higher charges and longer maturities is positive, as is the borrower’s trade-off to achieve a predetermined grant element benefit, but the slope of the latter’s indifference curve is steeper. This reflects the fact that a one-year increase in the final maturity of a loan will affect the grant element of the borrower relatively more than that of the lender because of the borrower’s relatively higher value of capital.

Chart 2Preferences of lenders and borrowers

Interest charge (In percent)

The lender is essentially indifferent between terms A and B shown in Chart 2, since the grant element cost of either loan is about 26 percent of the face value. The borrower, by contrast, would much prefer loan terms B, however, in which his grant element benefit is 45 percent, than loan terms A, which yield him a grant benefit of 38 percent. Clearly, there exist optimal terms such that neither the lender’s nor borrower’s position can be improved without cost to the other participant in the transaction.

A conceptual discussion of terms assumes, of course, that both donors and recipients have well-defined preferences. In fact, most developing countries have been price-takers and volume-maximizers. Donors have, by and large, not shown any inclination to appropriate a larger volume of aid at harder terms, an outcome that could, in all likelihood, be beneficial for borrowers. Legislators in particular seem inordinately concerned with the expenditure side of the ledger and insufficiently sensitive to the revenue aspects of foreign assistance. This absence of a present value focus is especially acute vis-á-vis IDA appropriations, which are financed by donors over the project lives of credits that stretch out ten years or more.

The potential gains that could emerge from a clarification of the preferences of donors and borrowers are all the more significant given the current shortage of long-term, low-interest financing. A hypothetical borrower whose long-term return on capital is 15 percent would gain a grant element benefit of 75 percent from an immediately disbursed loan at a final maturity of 40 years, with a 10-year grace period, and 3 percent interest. The grant element cost to the lender, using a cost of capital of 10 percent, would be 60 percent. Under these circumstances, a hardening of terms to those noted in the table as Terms B (final maturity of 20 years, 10 years of grace, and 6 percent interest) would enable the borrower to receive a one-third increase in grant-equivalent funds at no additional cost to the lender.

The table shows another example in which both the donor and the recipient should be indifferent between Terms A and Terms C. The former provides a longer final maturity with a ten-year grace period, while the latter would involve a relatively shorter amortization period but a longer grace period. While at the conceptual level the borrower is indifferent, it is possible that short-term liquidity considerations, such as those currently prevalent among major debtor nations, would distort the long-term return on capital calculation. Indeed, the specter of default might make it considerably more useful for a borrower to delay repayments on new loans (through a longer grace period). Theoretically, this should show up in a significant increase in the borrower’s discount rate, but his present-day preference may be almost infinite in extreme cases of illiquidity. Terms that are tailored to match these circumstances would be of significant assistance to debtors confronting a serious cash-flow problem.

Increasing the volume of assistance: an Illustration
Discount rate1Borrower2

(15%)
Lender3

(10%)
Terms A: 40–10–3GEB = 75%GEC = 60%
Volume A: 100Grant = 75Grant = 60
Terms B: 20–10–6GEB = 75%GEC = 30%
Volume B: 200Grant = 100Grant = 60
Improving the cash flow of assistance
Discount rate1Borrower2

(15%)
Lender3

(10%)
Terms A: 40–10–3GEB = 75%GEC = 60%
Volume A: 100Grant = 75Grant = 60
Terms C: 30–15–3GEB = 75%GEC = 60%
Volume C: 100Grant = 75Grant = 60

In the notation of terms, the first number is the final maturity of the loan (in years); the second is the grace period (in years); and the third is the interest rate (in percent).

GEB = grant element benefit.

GEC = grant element cost.

In the notation of terms, the first number is the final maturity of the loan (in years); the second is the grace period (in years); and the third is the interest rate (in percent).

GEB = grant element benefit.

GEC = grant element cost.

Policy issues

As already discussed, the discount rate is one of the major elements that determines the concessionality of assistance. Traditionally, a concessional loan has been defined as one with a grant element of at least 25 percent, using a discount rate of 10 percent. When such calculations were first made (by the OECD) some 16 years ago, the discount rate of 10 percent was a notional figure that was adopted as a working basis for calculation. Whereas in the earlier years it overstated economic rates of return in donor countries, in recent years it has understated considerably the opportunity cost of capital almost everywhere. But (given the generally accepted definition of official development assistance) if the discount rate had been adjusted to reflect prevailing interest rates, the concessionality of aid given as well as the total amount of assistance qualifying as ODA would have risen. This would have had an impact on perceptions of the donor countries’ aid effort (a United Nations’ sponsored aid-to-GNP ratio of 0.7 percent is the most generally used yardstick), but at the time it would call into doubt the practice of counting all loans with a concessional element above 25 percent as ODA.

Leaving aside the issue of the discount rate per se, current practice would allow two donors to receive equal credit in international measures of aid burden-sharing for similar amounts of ODA, although one donor might be offering terms that entail a grant cost of 26 percent and the other a grant cost of 99 percent. This bias could be eliminated by using a concept of “grant equivalent ODA,” whereby a donor giving $100 in grant aid and $100 in concessional assistance with a grant element cost of 25 percent would be credited with $125 in grant-equivalent aid. While solving one problem, this concept would exacerbate another, however, insofar as it would make the selection of the discount rate more problematic, and more political. Any average selected would benefit those donors whose national rates of return on capital fell short of the average, and vice versa. The alternative of using individual national discount rates, as was suggested but rejected by the OECD in 1967, would add precision but would make intertemporal comparisons difficult, particularly in light of recently volatile interest rates.

While it may prove impractical to calculate the grant equivalent benefit with great precision, and it may likewise be difficult to use individual country discount rates to calculate the grant equivalent cost to donors, the international community can certainly improve upon its concessionality measures. A first step would be to accept the fact that there are two divergent discount rates involved in measuring the value of concessionality. The second step would be to improve upon the conventional 10 percent rule for donors. But the major innovation would be to illuminate the different benefits accruing to recipients from alternative sets of terms, with the hope of increasing the potential value of a form of assistance that is increasingly difficult to expand in volume terms.

Other Resources Citing This Publication