Mismeasuring foreign aid

By Simon Scott and Hedwig Riegler^*

For several years now, CSOs have become increasingly concerned that donors are over-reporting their foreign aid efforts.

The root of the problem is that donors set the rules themselves in the OECD’s Development Assistance Committee (DAC). The key measure is that of Official Development Assistance (ODA), for which the United Nations has a long-standing target of 0.7% of donors’ national income. ODA is supposed to be concessional – to give something of value away.

DAC members as a group have never got near meeting the ODA target, but for political reasons they have also never dared to abandon it. Their unfortunate solution to the conundrum has been a series of dubious rule changes that weaken ODA’s concessional nature and therefore help donors look as if they are spending more without really doing so.

Controversy on some ODA items is not new, but a major turning point came in 2014, when the DAC agreed to change how it measured loans. From 2019 reporting on 2018 loans, ODA would no longer measure a loan’s actual disbursements and repayments, but only its “grant equivalent” – the amount it effectively gave away compared to a loan at market rates.

This was plausible in principle, but only if the DAC used realistic benchmark interest rates to estimate loans’ grant equivalents. Instead, it set the benchmarks so high that even loans at commercial terms would easily score as ODA. The Center for Global Development recently calculated that, as a result, total loan ODA reported by the DAC is now more than double the loans’ true grant equivalents.

The DAC justified its over-generous benchmarks by saying that they allowed upfront for the risk that loans would not be repaid. As Eurodad observed, the “rules reward donors for the possibility of future debt relief as soon as the loan is granted”. But in a remarkable about-face, in 2020 the DAC decided to count actual debt relief as an additional ODA effort. This clearly broke its 2014 promise not to double-count loan risk. The campaigning group ONE condemned it as “an unfair change, particularly in the middle of a global pandemic when more aid — not double counted aid — is needed to tackle COVID-19 and its aftershocks.”

Unfortunately, this is not the end of the DAC’s sly changes to the rules. It now also counts equity investments in developing countries as ODA, but with an over-generous twist. When donors eventually sell their investments, they will only subtract what they receive if it is less than the original investment: if it is more, they will disregard their profits, and report that they only received what they originally invested. This will systematically inflate ODA figures on equity investment.

The DAC also now allows its members a choice of two different methods for reporting the operations of their “development finance institutions”. These organizations are supposed to lend and invest in developing countries on a commercial or near-commercial basis. As no real subsidy is involved, their operations should not count as ODA at all. But donors have now given themselves two options for doing so: they can record how much they subsidize the institutions, or what the institutions disburse, and they can also switch between these two figures.

So far, all these changes have mainly distorted the figures of donors who give a lot of loans. Japan’s 2018 net ODA would have been only $10 billion on the old basis, whereas it scored $14 billion under the new rules. But the full impact of the changes will take years to be felt as donors gradually replace real aid with more and more commercial lending and investing.

In the meantime, quite a few developing countries have been accumulating unsustainable debts, and a wave of debt relief initiatives is in the offing. The new rules will ensure this is also massively over counted. One remarkable innovation is to allow donors to count more new ODA for forgiving an ODA loan than the borrower still owes. On private loans, DAC members kept an old loophole allowing them to score the whole amount forgiven, even though they only own the credit because the private lenders insured it with them against payment of market-based premia, and had to hand over title in order to claim their payout when the borrower defaulted. In most such cases, the premia accumulated are used – and more than suffice – to cover any loss. That means that no official budget funds are used for compensation payments – an offence against the ODA concept, which claims to measure official budgetary effort.

CSOs view the degeneration of DAC ODA statistics with increasing concern. In April 2021, three Austrian organizations – the Austrian Bishops’ Conference, the Austrian Foundation for Development Research, and Global Responsibility – organized a workshop to analyse the problems and discuss potential remedial action under “Chatham House rules”. This helped identify the root causes of the problem, and point the way towards a solution.

The workshop found that much of the trouble had come from the nature of the DAC as a forum for political negotiations. As the DAC Chair admitted in 2019, “political compromises are always and inevitably a bit messy”. Of course, even messy political compromises can serve useful diplomatic purposes, but they cannot produce credible and consistent statistics. For this reason, the OECD itself recommends that statisticians have “exclusive authority…to decide on statistical methods and dissemination…[and] are protected…from political and other interference in developing, compiling and disseminating official statistics”.

In the long run, therefore, the DAC will need to cede its ODA rule-making authority to a body composed of statisticians. Former senior OECD officials have recommended that these should include developing country representatives. Bringing in UN and other multilateral representatives as well, and allowing a greater voice to civil society, would help ensure that ODA statistics meet user needs and contribute to the international statistical architecture.

In the short run, the DAC should stop making politically-driven, piecemeal decisions on ODA rules. These now show clear evidence of a “ratchet effect”, where each new rule change over counts more than the last. Before any further distortions are introduced, the OECD should itself institute a thorough quality review in line with its own standards. At a time when developing countries urgently need real aid to recover from the COVID crisis, the international community must have confidence that official aid data are honest and reliable.