Development finance standards

OECD methodology for calculating imputed multilateral ODA


General issues

Different methodologies are possible for imputing aid by multilateral bodies back to the funders of those bodies.  In particular, the approach will vary depending on whether the intention is to show the share of the receipts of a given recipient attributable to a particular donor, or the share of a given donor’s outflows that can be assigned to an individual recipient. As DAC statistics are primarily designed to measure donor effort, the second approach is the one taken in DAC statistical presentations.


It is important to note that any methodology for imputing multilateral flows can only ever be an approximation.  Multilateral flows in a given year are not exactly imputable to donors’ contributions in that year, no matter what methodology is used.  Among the reasons for this are:

  • Delays. Donors’ contributions to international financial institutions (IFIs) are reportable in full on lodgment of promissory notes.  It may be up to three years or more before these funds are drawn down by the institution.  Even for institutions with annual subscriptions, there may be substantial delay between receipt of funds and spending them.
  • Turning grants into loans. The IFIs mainly lend rather than grant funds. Their lending in any one year is much greater than donors’ contributions, since it also draws on reflows of principal, interest receipts, and transfers of funds within the IFIs (net income from non-concessional lending is transferred to concessional lending).
  • Restrictions on pooling. In theory, core resources of multilateral agencies are pooled, so that all contributions can be considered to have contributed proportionally to all activities. But in practice, earmarks and restrictions cloud this simple correspondence between receipts and expenditure.
  • Limits of data on multilateral outflows. It is impracticable to maintain data on the outflows of all multilateral agencies, funds and programmes. The imputation methodology can only be applied to the finances of agencies for which both inflow and outflow data are available. 


The OECD’s calculation


1. The percentage of each multilateral agency’s total annual gross disbursements that each recipient country receives is calculated. This calculation is carried out only in respect of agencies’ disbursements of grants or concessional (ODA) loans from core resources.


2. The recipient percentages derived in step 1) are multiplied by a donor's contribution in the same year to the core resources of the agency concerned to arrive at the imputed flow from that donor to each recipient.  (Example:  In a given year, WFP provides 10% of its disbursements from core resources to Sudan.  Donor A contributes USD 50 million to WFP core resources in the same year.  Donor A’s imputed multilateral ODA to Sudan through WFP is 0.1*50million = USD 5 million).  This calculation is repeated for each multilateral agency.


3.The results from step 2) for all agencies are summed to obtain the total imputed multilateral aid from each donor to each recipient country.  


In practice, imputed multilateral percentages are calculated for about 20 agencies per year.  These account for about 90% of donors’ multilateral ODA.  Core contributions to the remaining agencies, for which the OECD does not have outflow data, are NOT imputed back to donors, so that imputed multilateral ODA remains slightly lower than donors’ total contributions to multilaterals.  Total imputed multilateral flows are not published on their own; they are shown in combination with bilateral ODA, in order to provide the most complete picture possible of the total ODA effort the donor makes in respect of individual recipient countries.


At present, there is no regular OECD's methodology for calculating sectoral imputed multilateral aid by sector or other aid parameters, though this has been done occasionally in the context of sectoral studies (e.g. on aid to the water sector, to basic social services, or in support of HIV/AIDS control).