Overcounting the ODA in loans
Although most ODA is provided through grants, a number of donors provide concessional loans.
In 2014, the DAC decided to change the way it reported the aid in loans. Instead of reporting the actual outflow and reflow of capital, it would instead work out the loan’s “grant equivalent”, and score that as ODA when the loan was disbursed. This change would be implemented starting with 2019 reporting on 2018 flows.
The DAC’s decision to use a net present value calculation to determine the amount being “given away” by a donor in extending a concessional loan is fundamentally a sound approach. But to assess the “grant equivalent” using this methodology, it is vital that discount rates are set at the right level. The higher the discount rate, the smaller the present value of repayments, and the higher the loan’s apparent “grant equivalent”. With a high enough discount rate, any loan – even one bearing an interest rate far above the rate a commercial lender would require – can appear to have a “grant equivalent”.
To measure the “grant element” in a loan, the discount rate should be set at the same level as the interest rate that the donor would need to charge to cover its costs, and no more. Using such a discount rate, any shortfall against the face value of the loan when assessing the present value of the borrower’s future repayments represents the cost to the donor. And the cost of finance for a donor borrowing in its own currency is determined by the current yield on its own government bonds at the relevant maturity. This is why the discount rates should reflect government bond yields.
However, instead, the DAC chose high discount rates – 6% a year for loans to upper-middle income countries, 7% to lower-middle income countries, and 9% to low income and “least-developed” countries, undifferentiated by currency, donor, or maturity. These rates comprised a “base rate” of 5% and “risk margins” of 1%, 2% or 4%, depending on the borrowing countries’ income level (which are unjustifiable – see here). In the very low international interest rate environment since 2014, discounting loans’ reflows at 6-9% p.a. generates hugely exaggerated “grant equivalents” for all major DAC lenders.
It’s not easy to say exactly how much “false ODA” is being scored, because costs of government borrowing constantly fluctuate, and they differ by country, currency and maturity. But there is no doubt that massive exaggeration has taken place.
One country, France, has even been boasting in its official budget papers that it scored more than five euros of ODA for each euro of real taxpayer subsidy on its loans in 2019 and 2020, and only slightly less in 2021. Page 11 here shows that it actually treats this ratio as a performance indicator for the efficiency of its management of ODA loans! Other estimates suggest that ODA scored for loans is 7-10 times their true “grant equivalents”.
The impacts of exaggerating the grant equivalent of loans are manifold.