When the negotiations for IDA17  were wrapped up in December, there was great relief that IDA  deputies were supportive of an IDA expansion despite their own significant budget difficulties. As part of that package, the World Bank Group itself pledged to give IDA $3 billion from profits.
This was a generous gesture by the World Bank (albeit a drop in the bucket of total aid), but how good was it for the global development effort? Consider the following—net disbursements of official grants and concessional loans (the category where IDA flows appear) have expanded from $39 billion per year in the 1980s (in constant 2005 dollars) to $85 billion in 2010 and 2011. In contrast, official non-concessional lending (the category where IBRD  and IFC  flows appear) has stayed steady. The latter was $15 billion in the 1980s and $22 billion in 2010/11. This picture is even more striking when considering the amounts in terms of recipient GDP. Grants and concessional flows to low income countries have gone from 3% of their GDP in the 1980s to 13% today, while non-concessional flows to lower middle-income countries (excluding India and China) have gone from 0.7% to 0.3% of their GDP. In fact, from 2000 to 2009, non-concessional flows to lower middle- income countries (and to developing countries as a whole) were negative, implying that developing countries repaid more to official development agencies than they received in gross disbursements.
The world has made a concerted effort to increase aid and this is good news for low income countries, but it is of scant help for middle income countries. Even though there are many more lower middle-income countries today than ever before, their aggregate share of grants and concessional assistance has remained the same. The pie simply has to be spread among more countries and more people.
The main agency responsible for helping lower middle-income countries is the IBRD. The IBRD was for many years the largest member of the World Bank Group. But in 2013 an astonishing landmark was passed—IBRD commitments were smaller than those of either the IFC or IDA. IBRD has become the “little sister” of the WBG after years of being the “big brother” that would always take care of its siblings.
So should we be applauding the transfer of large resources from IBRD to IDA?
This is a complicated question with no easy answer, and I don’t want to try and be comprehensive here. IDA needs to be large and strong to play a leadership role in a world where there are many more providers of aid and growing fragmentation. But the same applies to IBRD. So the question about taking from one to give to the other should at least be asked and seriously studied. A back-of-the envelope calculation shows what is at stake.
Let’s assume that global development is best served by trying to maximize the grant element of resources that are transferred to developing countries (I know that there are issues about aid effectiveness, but leave these aside for now). For each dollar transferred from IBRD to IDA, IDA can provide one additional dollar in grants.
But if the money was kept within the IBRD to build up its equity, it could lend more. IBRD is a leveraged institution, with an equity:loan ratio of around 27% in 2013 . This means it can lend out about $4 for each dollar in retained profits. Today, one of the most common instruments of IBRD lending is a 12 year, floating interest rate loan. This loan currently has an interest rate of 0.618%, and a front-end fee of 0.25%. At a discount rate of 10 percent (to approximate what lower middle-income countries might have to pay in commercial capital markets on their own, that is the opportunity cost of funds to them), the grant element of the loan is about 40 percent. This means that each dollar of capital invested in IBRD could generate 1.6 dollars of grant equivalence to developing countries.
So the mystery is this. Why isn’t there more discussion about how to increase IBRD’s capital and lending, using official resources? (Only a few, like Devesh Kapur and Scott Morris  are raising the issue.) Is the world better off if IBRD profits go to increase IDA, an unleveraged institution, or if they go to increase IBRD loans instead? In many cases, the client countries would be identical—there are 18 “blend” countries that are eligible for IDA credits and IBRD loans. And the grant equivalence could be raised by 60 percent if the money were kept in IBRD instead of being transferred to IDA!
Of course, in an ideal world, development advocates like myself would argue for both more IDA and more IBRD. Yes, there was a direct trade-off in the IDA negotiations, but in the overall picture of financing for development the amounts were small. But now that IDA17 is complete, we should shift attention to how to increase IBRD lending. There are plenty of ideas that don’t require a capital increase and the unpleasant budget politics that would entail. The World Bank’s Board has approved several proposals to enhance revenues and capacity: cost-savings, fee increases, and greater leverage can add significantly to the Bank’s ability to lend. Now it must make sure that it does not discourage countries from borrowing by unnecessary procedures and time delays.