G-20 finance ministers established an Eminent Persons Group on Global Financial Governance last April to review “the challenges and opportunities facing the international financial and monetary system” with a central focus on the multilateral development bank (MDB) system.

In a recent paper, prepared jointly with the Center for Global Development and the Overseas Development Institute at the behest of the Eminent Persons Group, we assess the role of the multilateral development system and the reforms needed to support the new global agenda. Unleashing the system’s potential will require greater coherence and political commitment across shareholders. The Eminent Persons Group will make their recommendations by October 2018, providing the last good opportunity to shape the future of the MDB system.

The milestone agreements of 2015 launched an ambitious global agenda to revitalize growth, deliver on the Sustainable Development Goals and take the necessary actions to mitigate climate change and adapt to its effects. While there is broad-based commitment to this agenda, progress on tangible actions has been lagging and the scale and urgency of the challenge insufficiently understood.

There is an urgency to the reforms. The coming decades will see the largest urban expansion in history, concentrated in emerging markets and developing countries. More infrastructure needs to be built in the next 15 years than the existing stock of infrastructure in the world. If this is not built in a sustainable way, it will become far more costly to retrofit to achieve low-carbon pathways, build resilience against shocks including climate change, and provide affordable access. It is also the last opportunity to address the demographic transitions occurring in the developing world, especially in Africa.  At the same time, major opportunities exist in the potential of new technologies and the capacity of the private sector.

MDBs are uniquely placed to play a central role in realizing the ambitions of the new global agenda. They can support policy and institutional reforms and build institutional capacity, enhance the quality of projects and programs, and scale up for transformative change. The unique financial structure of the MDBs allows them to leverage contributions from their shareholders and multiply them into financing at low cost, and use this financial capacity in turn to crowd in financing from other sources. MDBs can also help tackle spillovers, most importantly in the areas of climate action, disease containment, conflict prevention, migration, and economic and financial instability. Despite these inherent strengths, MDBs are constrained by their financial and institutional capacities, effectiveness of instruments, and unclear mandates and governance shortcomings. What can be done?

First, the concept of “graduation” based on per capita income needs to be adjusted. MDBs can add value across all income groups and drive the frontiers of growth and development where financial markets are imperfect and where knowledge and capacity gaps exist. A more adaptive framework would determine MDB engagement not just on income level, but on the development impact on the country and sector and the value of the engagement to the system as a whole.

While MDBs could do more everywhere, the multilateral development system could do far more in three underserved client groups: (i) fragile states; (ii) high-debt countries; and (iii) upper middle-income countries.

Fragile states cannot put in place the full array of policy and institutional arrangements of non-fragile states. Frameworks need to be tailored to local conditions and implementation capacities. The multilateral system needs to be more flexible in its operations in these places.

High-debt countries are often avoided by the multilateral system even though there is a risk of a vicious cycle: If no one is prepared to invest in infrastructure in high-debt countries, growth cannot recover and debt becomes relatively higher over time. The issue is the definition of the threshold for “high-debt” and the ability of countries to manage their debt obligations. We recommend that the multilateral system prioritize and strengthen debt management capabilities, especially when taking on large amounts of private debt to finance infrastructure.

Upper middle-income countries are often the source of innovation in financing and a conduit for learning. There are pressures for the multilateral system to disengage from upper middle-income countries. However, such a retreat would put greater pressure on business models (profits are reliably made from activities in these countries),  and would have the unintended effect of stripping countries and multilateral institutions of accountability at a global level. If China, Russia, and Brazil are not included, should other developing countries be held to high global standards on sustainability? How can multilaterals be held accountable for global results if they are not engaged in the largest developing country economies?

Second, MDBs individually and collectively need to become far more effective in unlocking private investment. MDBs are looking for ways to improve their catalytic role in mobilizing private financing. The G-20 Hamburg principles for MDB financing based on “reform, investment, and catalyzation” as laid out in the World Bank Group’s cascade approach is a good starting point. MDBs now need to improve their instruments and platforms for risk sharing and for mobilizing private investment and finance.

Third, there is scope for enhancing the effectiveness of the MDB system as a whole. Individual MDBs each perform well, according to evaluation results, but their impact could be enhanced if they cooperated more as a system.

One example is in infrastructure where multilaterals still principally deliver bespoke projects that are hard to scale up because they are demanding and time-consuming to prepare. But multilaterals are starting to collaborate more to build such systems. They have been key drivers on a multi-stakeholder initiative, SOURCE, whose website explains that it “provides support to national and subnational governments and public agencies in improving infrastructure project bankability, quality and delivery, in increasing investment and crowding-in private finance, in strengthening their technical capacities and abilities to manage risks.”

Locally, multilaterals are supporting client governments to develop institutions that will take on project planning, and preparation, as well as problem-solving in implementation. These national platforms permit financing from several different sources (including local currency long-term capital) to be pooled for a number of different projects.

Through these global and national efforts, multilaterals are helping develop platforms that can be scaled up. A noteworthy feature of current platforms is that they mobilize both multilateral and private capital as complements. The multilateral involvement assures the private sector of adequate due diligence in the platform approach, as well as permitting an apportionment of risk to different parties according to their ability to manage it. The participation of the private sector permits governments to achieve more.

Fourth, shareholders need to reach consensus on capital and governance to unlock the financial potential and effectiveness of the MDB system. The platform approach will require a sharp expansion in multilateral activity. Such an expansion is not feasible under current business and financial models that tend to hem MDBs in (for example, the use of paid-in capital and shareholder mandates are a disincentive to innovation). Research suggests that institutions with greater developing country voice have greater risk tolerance, while those dominated by advanced economies are more cautious financially.

One option for doing more is to encourage multilateral institutions to divest themselves of loans once projects have been built and start to yield a steady stream of revenues. After construction is completed, project risk is substantially lower and private capital should be more affordable. More rapid asset turnover by multilateral institutions would allow for engagement in many more projects.

The bottom line is that multilateral institutions are the best mechanism today for leveraging public resources. By our calculations, each dollar of paid-in capital could reasonably translate into $50 of public investments if properly allocated.

The current reform of the multilateral development system typically asks “are multilateral institutions doing the right things and, if so, should they be doing more?” By and large, the answer is “yes.” But when a different question is posed, “are we making sufficient progress on issues where the multilateral system can make a difference?” the answer is invariably “no.”

With so many issues on the table, we believe that the greatest risk to the system lies in active inertia, a tendency to make changes on the margin that fall short of a collective response that is scaled to the task at hand. For this reason, we recommend that there be periodic structured discussions on the performance of the multilateral development system in three fora: at the Development Committee of the World Bank, the United Nations’ quadrennial High Level Political Forum Summit on the Sustainable Development Goals, and the G-20. This may seem messy, but absent discussion and agreement in multiple places, little will change. The world cannot afford not to get the most out of the multilateral development system.