How can Multilateral Development Banks (MDBs) maximise the impact of their financial resources?
The role of MDBs in achieving global Sustainable Development Goals has steadily grown over the years, reaching $800 billion of loans in developing countries. Today’s overlapping crises make them even more necessary, but the exact way in which their resources should be mobilised is certainly giving rise to heated debates. In the 2010s, MDBs were seen as a way to mobilise private finance towards development projects, thus focusing on “derisking” investments and aligning private finance to sustainable objectives. While some experiments may have worked out successfully, we are still nowhere near the promised “billions to trillions” MDB contribution to financing for development.
An alternative approach seems in the making : last month,notwithstanding a certain amount of drama, the G20 Indonesian Presidency released an independent review on MDBs’ Capital Adequacy Frameworks (CAF). The report provides recommendations on leveraging MDBs further while retaining their triple-A rating, thus keeping their ability to borrow at low interest rates on global markets. Written by a group of experts, including several former MDB officials, the report proposes a new, rather radical goal: raising MDB lending capacity “several hundred billion dollars” in the long run, with a suggested implementation start date in the next 12 to 24 months.
The objective of this blog post is to provide some context MDBs and how their role has evolved. Some key points are:
- In the post-Global Financial Crisis environment, with relatively abundant development finance, MDBs were able to grow their role. They have become even more obvious in the current crisis and proposals to improve Capital Adequacy Frameworks are an important step forward. One risk that needs mitigating is that shareholders may see it as a substitute to raising more resources.
- An important implicit point of the CAF report is that it changes the approach to private finance mobilisation. The “cascade” or “blended finance” approach did not meet initial expectations, and channelling private finance through MDB’s balance sheets could be much more potent quantitatively.
- The report proposes a clear path to implementation, based on existing experiments. On some aspects, however, it remains timid. In particular, more thinking is required regarding the approach to liquidity regulations.
- Finally, one should not believe that the adoption of such recommendations would only imply changes at a few desks in isolated risk management units and Credit Rating Agencies. Leveraging more would most certainly require significant change in the kind of projects MDBS invest in as well as the kind of countries they work with.
Revising Capital Adequacy Frameworks should not come at the cost of lower investment in MDBs
The role of Multilateral Development Banks has significantly evolved over the past 20 years. The 13 largest MDBs have about $800 billion in aggregate development assets in develpoing countries. Of these $800 billion, about $700 billion are lent to governments, and the remainder to the private sector. Raising MDBs’ aggregate balance sheets by “several hundred billion dollars” would therefore mark a major shift in the global development finance architecture.
It would accelerate an existing trend: MDBs lending has progressively increased in the past decade, from $400 billion in 2010 to $700 billion in 2020, mainly concentrated in Lower Middle-Income Countries (LMICs) (Chart 1).
In relative terms, the importance of MDBs as a share of funding received by developing countries has declined: the rise of China and bond markets may have eroded their role, but the observation mostly applies to Upper Middle Income countries. In contrast, their lending portfolio for LMICs has only slightly declined (from 35% of total borrowing in 2010 to about 28% in 2020), and has remained broadly stable for Low Income Countries (LICs), at 40% of their total borrowings.
What will the future hold? The growing importance of MDBs was made possible thanks to multiple capital increases, notably at the Asian Development Bank in 2009 (by 200%), the International Bank for Reconstruction and Development in 2018 (more than doubling its equity), the African Development Bank in 2019 (by 125%), and multiple International Development Association replenishments.
What the CAF report does is that it offers a new proposal to keep growing the role of MDBs: in addition to new capital infusion, MDBs can leverage more in order to increase their lending headroom. However, some might be concerned that such a strategy would not come in addition to shareholder support, but as an alternative. Indeed, one should not forget that capital increases require complex and protracted negotiation processes (the last IBRD General Capital Increase, was, for instance, linked with data manipulation of the Doing Business report). Could Balance Sheet Optimisation be a convenient way to avoid, or at least, delay, new capital increases?
This should not be the case: the sources that supported development finance in the last decades are likely to slow down. The Belt and Road Initiative is in a relative standstill, other bilateral partners such as the OECD Development Assistance Committee (DAC) members are reducing or maintaining their support, with possible displacement effects from the war in Ukraine. It is unlikely that private sector lending will return to its levels of the past decade. MDBs should provide low-cost finance to their clients, and accelerate the role of their balance sheet.
2. Is there a blending vs. leverage trade-off?
The CAF report has also a more implicit premise, which could mark an important shift in the thinking around private finance mobilisation for development. The latter has become the lynchpin of financing strategy since the mid-2010s. World Bank “Cascade” or “Maximising Finance for Development” (MFD) strategies sought to leverage sources of official finance, often concessional, in order to raise private resources. This approach has mostly failed to materialise: leverage effects are small, often below 1 (there are exceptions, such as policy-based guarantees, which tend to have leverage effects between 1 and 3) whilst the share of sustainability-oriented funds in developing countries remains tiny. They are also procyclical: private finance mobilized fell sharply during the COVID-19 crisis, while MDBs funding peaked.
Private finance to developing countries, especially in Low and Lower Middle-Income countries, is expensive, whether under the form of project-finance, or loans to governments. MDBs have been remarkable in obtaining much lower rates and passing them to their clients. Borrowing through MDB-issued bonds allows for larger volumes of mobilization, reaching the scale that blended finance projects have struggled to reach. Note that these two approaches do not necessarily contradict each other – they are aimed at different kinds of investors. However, there are some trade-offs to consider. Increasing leverage might not imply a loss of the treasured AAA rating, but it might raise, all else equal, the cost of funds for MDBs. To compensate, MDBs will need to increase the quality of their assets, and thus reduce the share of riskier investments often associated with blending (first-loss provisions, mezzanine tranches, etc.).
These trade-offs will depend on exactly how balance sheet expansion is achieved, but a shift in strategy and emphasis is very likely.
3. A smart strategic approach grounded in existing experimentation
Third, the value of the report is to provide a clear path to implementation. One key constraint to implement such solutions is the lack of harmonised approaches to risk appetite by MDBs. Internal risk assessment models by MDBs tend to be conservative and ignore callable capital in their financial decisions. In addition, the three main Credit Rating Agencies apply different methodologies, which means that to retain their AAA rating, MDBs have to satisfy the most conservative approach.
The report provides proposals on achieving such mobilisation and divides them into two categories:
- Allowing MDBs to lend more by better taking into account their business models (in particular, giving more credit to callable capital, integrating their preferred creditor status in internal risk assessments and Credit Rating Agencies methodologies; and relieving the constraint linked to the concentration of their portfolio). This would essentially allow MDBs to better leverage their existing equity.
- Experimenting with risk mitigation measures, including by securitising claims, offering various forms of non-voting capital, etc. where private lenders could directly finance some operations, especially for non-sovereign loans. A clear conclusion of the report is that some of those experimentations are already taking place in some MDBs, and that simply taking stock of the successes and adapting them to the larger MDBs could yield gains in lending headroom.
One area where the report falls short of providing such specific advice is on the models of liquidity and funding. It is well known that MDBs are required to hold large amounts of liquid assets: on average, for every three dollars lent to developing countries, one dollar is kept in ultra-liquid AA rated asset. This is due to the fact that unlike traditional commercial banks, MDBs do not have access to a liquidity backstop. They therefore have to maintain enough liquid assets to meet their short-term obligations, often defined as debt with maturity due in the next 12 months, plus next-year committed disbursements.
The report does not dive deep on this topic, but the same criticisms addressed to risk-weighting of assets could be replicated for the approach to liquidity. CRAs assessments are not coherent, leading to an alignment to the most conservative, especially considering that MDBs are often more conservative than what their own guidelines would imply. Finding ways to provide a liquidity backstop, for example through pooled provision of a line of credit, would complement efforts on capital adequacy: for an increase of $100 billion in balance sheet, the IBRD retains $30 billion in “safe” assets with no development impact, while the EIB, thanks to its ECB backstop, only retains $20 billion.
4. Beyond financial models, more leverage will probably imply drastic operational evolutions
The report rightly focuses on the issue of capital adequacy – but the implications of such a sea change are much broader. If MDBs increase in their size while retaining AAA rating, they will need to compensate by reducing risks on the asset side. By implication, this probably imposes to focus on countries with relatively high credit ratings, i.e. a subset of middle-income countries which do not have major debt sustainability problems. Low-income countries will continue to rely on grants and concessional loans, which allow less leverage, even with a new capital adequacy approach.
Is it feasible in the context of high risk to sovereign debt? In addition to raising its level of lending, MDBs would probably need to change the composition of their portfolios. While the 1990s and the 2000s saw a marked increase across the board of MDB’s client GDP per capita, the trend has stalled and reversed in many cases. There are many explanations: sluggish growth in middle-income countries such as Brazil and Mexico, but also a matter of strategy to focus on countries with less market access, for IBRD for instance. As a result, most MDBs lend to countries which are on average poorer than in 2010.
Such an evolution would have to be reversed, and operations could be geared back towards upper middle-income countries, with a focus on climate mitigation and more generally the Global Public Goods agenda. In a recent column, Rabah Arezki and Philippe Le Houerou proposed such an evolution, which could match well with more MDB leverage. This is an area with large investment needs, which makes sense, but it will also require organisational changes well-beyond finances, and adapting operational levels throughout organisations.
At the same time, this reform will improve sustainability profiles for developing countries. Substituting MDB funding to private loans will improve the debt profile of borrowing countries, by lowering interest rates from a current 7-10% for most emerging markets to 3-5% currently paid on non-concessional loans.
As far as countries are concerned, higher debt to MDBs will also mean less private borrowing: MDBs tend to set non-concessional borrowing limits, which would be more likely to be binding if their own lending were to rise significantly. In addition to shifting back to relatively richer countries, the private-sector arms of MDBs are likely to be the ones that will benefit most from leveraging balance sheets.
There will be a lot of attention on next steps of a possible implementation of the recommendations. Already, Fitch has expressed its reserves towards the possibility of downgrades if MDBs were to expand their lending ratios significantly. Ambitious reforms will require a lot of concertations among shareholders, MDBs themselves, markets and Credit Rating Agencies.
MDBs have been essential in mitigating the impact of the COVID-19 crisis, and their ability to finance the recovery to the current shock and long-term needs for climate change hinges on optimizing their financial model.
This note benefitted from insightful comments and discussions by Rabah Arezki, Hamouda Chekir, Daniel Cohen and Ishac Diwan.
 The report reaches a total of $1.3 trillion, but this includes the European Investment Bank’s balance sheet of $552 billion, of which the large majority is held in the European Union or the UK. Only about $40 billion is held in developing countries.
 In addition, MDBs provide about $4 billion of grants annually, which rose to $8 billion in 2020 due to the COVID-19 crisis, predominantly to LICs.
 With a decline in 2010-15, which was reversed after 2016.
 Hans-Peter Lankes (2021), “Blended finance for scaling up climate and nature investments", One Planet Summit report.