- Issue Briefs and Special Reports
- Jan 31 2018
The global effort to meet the targets set by the Paris Agreement on Climate Change and the Sustainable Development Goals will depend crucially on reforming the structure of development finance. Mobilising private capital will be an essential part of this effort, and existing development finance institutions, led by the complex of multilateral development banks (MDBs), will have to re-orient their strategies and functioning to prioritise this mobilisation. While a rhetorical commitment to this re-orientation has been made, there remain significant functional, operational, and geo-political hurdles to the transition. Greater focus on developing-country priorities, willingness to handle and moderate risk, and the transformation of local operations are essential if MDBs are to stay relevant and effective.
International development finance architecture has long been underpinned by an interlocking system of multilateral development banks (MDBs) that were conceived of as one of the primary conduits for fund flows from the developed to the developing world. Over time, these MDBs acquired a disproportionate level of influence on developing-world economic policy choices, while failing to sufficiently democratise their governance structures. However, as the human resource pool from which the MDBs drew their decision-makers eventually became more democratised, they grew more responsive to the requirements of the developing world.
The threat to MDBs' functioning, however, has multiplied in recent years, driven by various processes that are examined in section one of this brief. While it is too soon to suggest that MDBs will be rendered archaic by the changing contours of global development finance, these growing pressures on traditional MDB functioning have led to some recalibration of the objectives of many MDBs. Some of these developments are examined in sections two and three.
The question, however, is the degree to which a new focus for the MDBs as catalysts for private finance in climate change-sensitive development infrastructure will be effective. There remain various obstacles for the MDBs, some of them beyond the MDBs’ control. Sections 2 and 3 of this brief provide an outline of these hurdles, and lay out the further study required of MDBs' objectives, strengths and threats going forward; Section 4 examines possible dangers in the MDBs' current reorientation to meet climate goals. The focus throughout the brief is on the political and politicised challenges that need to be overcome to enable the transformation of MDBs into suitable instruments for energising cross-border capital flows into green infrastructure in such a way that Paris Agreement targets are met alongside each country's development goals.
I. A Low-effort Equilibrium
Multilateral development banks (MDBs)—most importantly the World Bank (WB) but also regional banks such as the Asian Development Bank (ADB), the African Development Bank (AfDB) and other development finance institutions (DFIs)—have long been the pillars of financing for global development.
As the decades passed, there were increasing demands that MDBs "democratise" their governance by giving developing countries a greater say in their operations. This pressure was moderated by various actions of the MDBs themselves. For example, they became more responsive to social-sector demands from non-governmental organisations (NGOs) in both developing and developed countries. (In India, the World Bank's approach to protests surrounding the Sardar Sarovar dam on the Narmada river, in which it was more responsive than state or central governments, is illustrative.) Meanwhile, the MDBs' catchment area for staff widened over the years, incorporating many members of developing-country elites. These staffers would informally represent the concerns of developing-country governing classes in internal MDB discussions, ensuring that MDBs made their functioning more broad-based to a degree, if not their formal governance structures.
Eventually, a comfortable equilibrium developed that satisfied all the various interest groups involved: the MDBs' financiers (developed-world governments), the MDBs' staff, and governments of countries receiving the MDBs’ funds.
In this equilibrium, MDBs focused on disbursing loans directly to public sector authorities in developing countries. These loans would be provided at concessionary rates, and developing-country governments used them for designated projects, which would also be monitored by MDB staff. All concerned were comfortable: MDB staff, because the process was easily manageable and required little specialist skills; the net donor states, because MDB boards could be satisfied that lending was suitably restrained, parsimonious, and did not benefit private-sector players directly or unduly; and developing-world governments, because they had control over the funds flowing into their countries from the MDB, and their own public-sector institutions were the primary beneficiaries. While the pitfalls in private financing for infrastructure, including through such models as public-private partnerships or PPPs, have become clearer in recent years, it is vital to keep in mind that public investment also has inherent problems of its own. In particular, it is wasteful of scarce resources; the International Monetary Fund (IMF) estimates "average inefficiencies" in public investment processes at around 30 percent.[i]
It is important to note that this is not in fact how the Bretton Woods DFIs were imagined at their inception. Christopher Humphrey and Annalisa Prizzon[ii] quote US Treasury Secretary Henry Morgenthau, who was an important figure at the time the post-War development consensus was being created, as envisioning a different primary task: "The primary aim of such an agency should be to encourage private capital to go abroad for productive investment by sharing the risks of private investors in large ventures ... The most important of the Bank’s operations will be to guarantee loans in order that investors may have a reasonable assurance of safety in placing their funds abroad." Yet it is clear that this is not, in fact, how things have turned out in practice. Humphrey and Prizzon point out that, in 2013, only 1.7 percent of the lending approved by DFIs took the form of various guarantees. Thus the first, increasingly pressing, problem: MDBs have become too comfortable with concessionary loans as a method of development finance.
The second problem that began to press over time was that MDBs were not living up to their initial commitment to finance infrastructure. According to Nancy Lee of the Centre for Global Development (CGD), the total commitments of MDBs (sovereign and non-sovereign) are about US$116 billion per year, of which infrastructure funding is only about $45 billion.[iii] This comes at a time when private financing for infrastructure is declining. The World Bank's annual report on private participation in infrastructure found that the commitment of resources with private participation in infrastructure in 2016 was the lowest in 10 years.[iv] From $210 billion in 2012, such investment had come down to just over $71 billion in 2016. The World Bank argued that this was driven in particular by steep declines in the number of projects being financed in three major emerging economies: India, Turkey and Brazil.
Nor is it likely that, given tightening monetary policy and increasing returns in the developed world, this pattern of shrinking will be easily reversed over the next few, crucial years. In other words, private financing of infrastructure was and is falling off, but MDBs were and are unable to step up and fill the gap—their lending seems as susceptible to business cycles as was private investment. Although the Inter-Agency Task Force on Financing for Development has claimed that MDBs were able to play a quick counter-cyclical role immediately after the global financial crisis of 2008, this was clearly neither sustained over time, nor properly directed.[v] Obviously, MDBs were not performing as designed.
These two problems became particularly potent barriers to MDB functioning as the scale of global poverty decreased in recent decades and state capacity and aspirations in the developing world increased. The binding constraint on further growth in these economies was the paucity of world-class infrastructure. MDBs and DFIs were clearly under-capitalised if they were to fill an infrastructure spending deficit that grew to $1-1.5 trillion a year. The ADB, for example, can produce a meagre $13 billion annually in new loans.[vi] Yet there is little or no appetite among the principal shareholders of existing MDBs to increase the capital available. Nor is it obvious that any such appetite will develop going forward; in fact, pressure is building domestically in many source nations—particularly the US— to focus on expensive infrastructure expansion locally.
The conclusion seems inescapable: the existing model of MDB activity is failing to adapt to the needs of the 21st century. To this existing crisis of MDB finance, two additional wrinkles have been added: growing concern about climate change, and the growth of investible surplus capital in the People's Republic of China.
The imperatives of climate change require cross-border infrastructure finance to not only consider the previous constraints on its operation such as currency fluctuations, sovereign risk, contract enforcement, and long tenures, but also to examine the sustainability of the assets so built and whether they feed into the broader attempt to control and respond to global warming. On one level, this means that there is an additional objective for MDBs to take into account, when they are already struggling with multiple, sometimes contradictory aims. Their lending is supposed to be safe, create broad economic externalities, avoid crowding out private investment, meet target country requirements, adhere to governance standards, and avoid alienating NGOs. Now the global consensus against carbon also has to feed into decision-making, creating an additional "co-benefit" that MDB credit has to address. Going forward, there are only two likely responses to this: paralysis or over-reaction. Paralysis is visible in the unwillingness to increase the capital on call for most MDBs, and over-reaction in pledges such as was recently on offer from the World Bank to stop any and all support of carbon-based upstream energy projects. The dangerous consequences of this will be examined in Section 3.
The emergence of the People’s Republic of China as a major player in development finance has both a constructive and a disruptive side to it. Increasing funding for "hard" infrastructure globally is a major political priority for the current leadership in Beijing, as seen from the centrality of the Belt and Road Initiative (BRI) in its foreign-policy messaging. For many observers, this is not a negative development. Indeed, most countries straining to attract capital into infrastructure are clearly willing to open themselves to the benefits that could, from an optimistic viewpoint, accrue from access to PRC funds. However, the disruptive effects are also worth considering, especially as they will make the transformation of existing MDBs more difficult.
For one, going forward, attempts at coordination between MDBs will be complicated by the geo-political competition underlying Beijing's creation of an alternative development finance architecture anchored on its own multilateral development bank, the Asian Infrastructure Investment Bank (AIIB). Second, concerns have understandably been expressed about a "race to the bottom" in terms of governance and political standards associated with MDB activity if the existing DFI complex begins to compete with the new Beijing-centred DFI complex as a source of finance.[vii] Third, the broad division between the two DFI ecosystems threatens any nascent cooperation and universalisation of standards, templates and databases relating to project finance.
II. The MDBs' New Aspirations
The leaders of MDBs and their shareholders have not been completely quiescent in the face of this growing challenge to the MDBs' role as the primary pillars of global development finance. In some cases, their responses have gone in the wrong direction, but in others, they have tried to realistically bridge the gap between what needs to be done and what can be done.
On the positive side, it is clear that a new consensus is growing around the definition of the MDBs' role. Several methods of closing the lacunae identified earlier in this brief are being examined. The MDBs' "joint declaration of aspirations" (JDA) in 2016 set targets for infrastructure lending, for example, that they subsequently declared were either close to being met or had been met. The declaration sought to refocus the MDBs' efforts towards infrastructure, by methods "including formulating quantitative ambitions for high-quality projects, encouraging multipartite cooperation financing models, catalysing private resources, fostering collaboration between new and existing MDBs, and strengthening project preparation to improve quality and bankability".[viii] By the end of 2016, the MDBs claimed that the quantitative ambitions at least were being met.[ix]
The other aspects of the MDBs' aspirations are less easily quantifiable and thus harder to evaluate. For one, consider the word "quality" before infrastructure in the above listing. "Quality" is defined in a particular way in the JDA document, with the first requirement being sustainability over the life-cycle of the infrastructure asset, including climate resilience and carbon mitigation. But the word, in the context of international infrastructure investment, also has a specific undertone: there are often concerns that Chinese-built infrastructure is sub-standard, and thus other larger builders and funders, especially the Japanese agencies, emphasise "quality" in their own pitch to developing countries. This is only one example of a possible geopolitical pitfall in the path to modernisation and coordination of MDBs' goals and operations.
Two specific directions of the MDBs' new focus require closer examination. The first focuses on the pipeline of new projects in the developing world, and consists of a move from MDBs' role of simply funding projects, to curation and risk mitigation; and the second is their stated ambition to effectively mobilise resources from the private sector, including in the developed world.
The first direction makes the reasonable assumption that there is suppressed demand in the global North, especially among institutional investors, for long-tenure investments with the appropriate risk-return profile. If institutional investors can help create a "pipeline" of such projects, through proper preparation, guarantees, or co-financing, this suppressed demand can help fill the $1-$1.5 trillion investment gap. The end-2016 report on how the MDBs have moved towards addressing the goals of the JDA, identifies a host of new attempts to create a project pipeline: "Therefore, in addition to MDBs' traditional portfolio of products for infrastructure development such as non-sovereign financing windows, guarantees and other co-financing and risk-mitigation instruments, and new specialised project preparation, the MDBs have come together to support the G20 Global Infrastructure Hub and the World Bank Group-hosted Global Infrastructure Facility, which will support greater collaboration in preparing and structuring complex infrastructure projects to attract long-term financing from private investors. The MDBs are strengthening the infrastructure pipeline through project preparation facilities (PPFs). These include the Inter-American Development Bank’s (IDB) InfraFund, AfDB’s New Partnership for Africa’s Development’s (NEPAD) Infrastructure PPF, European Investment Bank (EIB) hosted initiatives such as the Arab Financing Facility Technical Assistance Fund (co-managed by Islamic Development Bank and IFC); European Bank of Reconstruction and Development’s (EBRD) Infrastructure PPF; ADB’s Asia Pacific PPF, as well as AfDB’s Africa50 Initiative, which will focus on both project preparation and project finance."[x]
The World Bank Group has rhetorically committed itself to a "cascade" approach to its operations, which one senior WB official described thus: "To better sequence our interventions, we’ve developed a ‘cascade approach’ to investment decision-making to encourage private sector participation, while leveraging and preserving scarce public dollars for critical public investments. If commercial financing is available, that is the preferred course. If it is absent, we try to address market failures. If those efforts are unsuccessful, we use risk instruments and our own matching capital to try to encourage private investment. Finally, if absolutely necessary, then public and concessional financing will be used."[xi] Some of this rhetorical commitment has been matched by the direction of MDB finance, with much of the $75-billion "18th replenishment" of World Bank funds for the poorest countries, or International Development Assistance 18 (IDA18), being used to set up a joint venture between three World Bank Group entities—the Multilateral Investment Guarantee Agency, the International Finance Corporation, and the International Development Association—that focuses on risk mitigation, guarantees, and blended finance.
Overall, this is a welcome development. The MDBs' extensive experience in most developing countries, their superior knowledge of local conditions and decision-makers, and their political heft all mean that they are in an excellent position to serve as mitigators of risk, or preparers of developing-world projects for private developed-world finance, or both. This is a vital new direction, and it has clearly been taken on-board at the strategic level by MDBs. But the implementation hurdles are also visible in the paragraph quoted above from the end-2016 inter-agency report. There are too many different and overlapping attempts, reflecting the multiple priorities of the DFIs involved. Information sharing between the MDBs is marginal; it is reportedly difficult even to get different silos of a single institution or group to share information about observed risks. Indeed, the inability to share information has always been a problem with MDBs, leading in the past to competitive subsidies, for example; but in an era in which MDBs are supposed to focus on risk mitigation, it takes on a new and sharper edge. In addition, pooling information and resources will enable better risk diversification. It is noteworthy that this happens even as many organisations in the existing DFI complex share the same influential shareholders.
The Global Infrastructure Hub under the G20 is one mechanism that is supposed to help overcome these issues, but the listing of projects on the Hub seems to depend entirely on the enthusiasm of individual decision-makers within specific national or local governments in the developing world. In December 2017, for example, only three projects from India were listed on the Hub, all of them from the Railways Ministry in the Union government. It is not yet seen as a core duty of any operational head of an MDB in a developing country to work with the different local agencies involved in supervising and bidding out infrastructure projects, and raise those project proposals to the level required for listing them on this proposed project pipeline. Moving from the strategic to the functional is proving to be a problem.
The conclusion is clear: while the MDBs' strategic commitment to altering their functioning may well be genuine, and is a step in the right direction, the operational impact of this strategic decision will be minimal unless it is followed up with specific actions to incentivise its staff.
III. Building a Closer Relationship with Private Capital
Similar barriers exist in the MDBs' attempts to move in the second and related direction – towards becoming catalysts for private finance. Acting to explore risk mitigation requires a close understanding of the nature of the destination countries for investment, but shifting focus to catalysing private investment also requires MDBs and DFIs to understand the incentives and requirements of private sector actors.
As with the replacement of risk mitigation by direct lending as the primary instrument for MDBs early on in their history, the distance that most DFIs maintain from private capital is the direct opposite of how they were originally envisaged. Humphrey and Prizzon point out that "the World Bank was viewed with considerable suspicion by the New York financial community" when it was first launched, which incentivised MDBs to move away from creating products to appeal to the financial markets. This dynamic has been intensified by the increasingly fast pace and specialised requirements of modern finance, to which the tightly controlled and slow-moving MDBs have been unable to adapt in most cases. One exception has been the world of guarantees for trade finance, which provide a useful model for the development of other products and services by MDBs that can appeal to the private sector while meeting destination-country development objectives. The European Bank for Reconstruction and Development (EBRD) introduced quick turnover approvals for trade finance guarantees—in many cases within two days—over two decades ago. Such approvals did not need to be taken up to the MDB's board for approval on a case-by-case basis. This solves at least one part of the problem that hinders cooperation between MDBs and private finance.
There is a stark difference in treatment between guarantees for trade finance and for other MDB focus areas. There continue to be major structural barriers to the growth of guarantees as a replacement for concessional loan finance as the method of choice for MDBs. Nancy Lee of CGD points out that guarantees are essentially treated identically to loans when it comes to accounting and provisioning at MDBs, although in fact they have a significantly lower actual risk. Meanwhile, guarantees involve higher levels of effort for MDB officials, since there are two parties to negotiate with: private capital, and the project promoter in the developing country. Given the equal book risk and the greater organisational input required, MDB staff have clear incentives to de-prioritise guarantees and stay within the existing, low-risk/low-outcome equilibrium. The instruments themselves as designed by the MDBs are considered too complex by investors, and MDB operations too bureaucratic; Lee quotes a Convergence study saying that only 12 percent of blended finance deals involve guarantees or insurance, and points out that an even lower proportion, below six percent, of World Bank Group deals involve guarantee structures or other risk management instruments.
Essentially, MDBs will have to accept that a part of their duty is to handhold private capital in geographies that the latter finds intimidating or complex. This will require willingness on the part of MDB boards to take on greater risk, as well as to accept that they are working not only for governments but also for private capital. The current arms-length relationship with private capital may be comforting and easier to manage or supervise, but it also means that MDBs are failing in their core aims of increasing the access to funds for a broad segment of projects and countries. In other words, in creating financial products that serve to catalyse the cross-border flow of funds to the sort of infrastructure projects required in the developing world, MDBs will not only have to take into account their own incentives and the overall requirements of the destination countries for the funds, but also accept that they are working to serve the profit motive of private capital. Their resources will not only underwrite but also ameliorate the risks being taken by private players—not an easy psychological transition for them to make. However, it is one that is deeply necessary. Observer Research Foundation’s work on the structure and sociology of Western institutional investment reveals that a major barrier to cross-border capital flows into climate-resilient infrastructure is the lack of expertise on other geographies that such investors have in-house.[xii] MDBs have such expertise; they will need to help expand the capacity of private capital to make the right choices in markets that these funds and investors find opaque.
In short, MDBs will have to move from lending to risk amelioration; and from amelioration to intermediation. They will have to serve, in fact, as real banks – intermediaries between the real pools of capital and the most productive and important destinations for those savings.
V. Meeting Local Priorities in Climate Action
The growing centrality to development finance of the carbon control consensus, while welcome on several levels, is nevertheless being operationalised in a manner that threatens not just the goals that the MDBs have set themselves, but also the spirit of the Paris Agreement and the broader fight for sustainable development. As an example, consider the declaration in early December 2017 that the World Bank will cease financing, from 2019, any and all projects related to oil and gas exploration and extraction. This is an example of exactly how poorly constructed the MDB response to the climate action agenda has been. Operationally, the climate agenda plays a negative rather than a positive role in MDB actions. Rather than as an effective stimulus to raise its direct financing of sustainable energy projects worldwide, it is a cause of constraints on financing projects that do not meet carbon mitigation parameters set in the global North.
How does such action fit into the structure of the 2015 Paris Agreement on climate change? The explicit reason that the Bank gave for this decision is that it was meant "to help countries" meet the targets that they had set as part of the Paris Agreement. Yet the underlying spirit of the Paris Agreement was clear: it was to allow sovereign nations to plan and implement their own paths to the needed controls on carbon emissions, while respecting the energy needs of their populations during the transition. Imposing on these sovereign choices by ending funding unilaterally is a clear violation of the spirit of the Agreement. It is also unnecessary; the focus of climate action should be on scrutinising and aiding the implementation of each country's Intended Nationally Determined Contributions (INDCs).
Finally, what will be the outcome of such unilateral action by existing MDBs? It could be argued, given recent history, that projects stranded by such decisions -- this particular decision might create stranded assets in Egypt and Mozambique -- will turn to alternative sources of development finance, and in particular the pool of investible capital controlled by China. In the absence of commitments by the Beijing complex of DFIs to climate-related monitoring of its investments, it is hard to see how exiting such projects will lead to an improvement in standards.
What will certainly be a negative going forward is if, as part of an over-reaction to the introduction of climate issues as co-benefits, MDBs exit from projects that create institutional capacity within countries to regulate, negotiate and scrutinise carbon-related projects. The December decision by the World Bank threatens a $29-million IDA credit to Senegal for this purpose, aimed to "ensure Senegalese oil and gas development projects take place in an environment inductive to private sector investments aligned with the public interest".[xiii] It strains credulity to suppose that extraction from any new discoveries of oil and gas resources off the coast of Senegal will in any way be affected by the lack of MDB funding for state capacity. In all likelihood, what will be affected instead will be the weight given to public concerns about sustainability, and to the fit between exploitation of these resources and Senegal's INDC.
A pattern of such actions will serve to reinforce the notion that MDCs are repeating in a more climate change-conscious age the errors many hoped they had begun to leave behind in the 2000s: namely, a refusal to consider the domestic priorities and concerns of destination nations. DFIs were close to defining themselves as sources for catalysing private investment into the projects determined as important by developing-world governments. These projects would be chosen keeping in mind those countries' commitments to the Paris Agreement and the Sustainable Development Goals. This important redefinition is under threat if an attempt to curry favour with some influential factions in donor countries distorts the macro choices made by the MDC governing complex.
If climate-resilient infrastructure is to be built that addresses the need to meet the Sustainable Development Goals and underwrites countries' efforts to keep their Paris Agreement timelines, the development finance architecture must change. Existing and future committed resources will not be enough if DFIs operate as they always have. Rather, a clear commitment to focus on creating a bridge between private capital and "quality" infrastructure in the developing world is essential. It is important and welcome that a rhetorical commitment to this re-orientation of the current DFI complex has been made.
However, this strategic commitment must be matched with an understanding of what else must change. In particular, it should be noted that the crucial gap in developing countries that prevents them from attracting private capital to their infrastructure projects at the scale required is the inability on their part to design, monitor, de-risk and independently evaluate projects. Filling in the information and capacity gaps is thus low-hanging fruit -- but it will require re-orientation of MDB operations on the ground, and not just of their mission statements. Changes will also be needed to the functioning at board level; they should not waste time on the evaluation and approval of individual projects, but on setting the overall parameters that middle-level officials can use. This will help reduce the "bureaucratic hurdles" that private investors see as a major stumbling block to doing business with DFIs.
Two other changes to the strategic direction of the existing DFI complex are required, if it is to stay within the spirit of the Paris Agreement and avoid geopolitical disruption. A subordination to local-government development objectives should be internalised at all levels of existing DFIs; if nothing else, this is a crucial step to de-risking any project and insulating it from local political currents. Second, given that additional capitalisation is unlikely at any scale, MDB boards should at least move away from their fear of risk and of close relations with private capital. Obviously, transparently safe lending is not why DFIs were set up. Nor is it in keeping with the new aspirations embodied in the 2016 JDA. MDBs need to restructure their human resources, their internal incentives, and their local operations to achieve a particular double-sided matching. The financial instruments they offer must subordinate themselves to developing-country development goals and priorities, and they must be designed to be accessible and simple enough to be taken up by private finance, and in close co-operation with the private sector. If this is done, the broader aims of building green infrastructure and meeting sustainable development goals will be met.
[iii] Lee, Nancy, “Billions to Trillions? Issues on the Role of Development Banks in Mobilizing Private Finance”, November 17, 2017.
[xi] Mohieldin, Mahmoud, “Remarks by World Bank Group Senior Vice President Mahmoud Mohieldin", CEO Summit - Sustainability Leadership Conference 2017, March 15, 2017,
[xiii] World Bank Group, World Bank to Help Senegal Negotiate Complex Oil & Gas Projects, May 31, 2017,