Mining for GEMs

High on the list of objectives in the ongoing multilateral development bank (MDB) reform debate is to increase levels of private capital mobilization—something that has proved much harder to achieve than expected. Despite the lofty ambitions set out in the “billions to trillions” agenda, each dollar of public investment has netted only sixty to seventy cents of private investment.

In weighing options to improve this record, MDB shareholders have coalesced around one proposal: providing open access to default and recovery data from MDBs and development finance institutions (DFIs) which are currently housed in the Global Emerging Markets (GEMs) Risk Database. Their rationale is that access to this data would enable private investors to better assess risk in countries where they may otherwise be reluctant to do business.

The G20 have endorsed the idea of a stand-alone database (“GEMs 2.0”) and called for its establishment by early 2024. However, prospects for meeting this timeframe appear slim because the process involves several complex issues that need to be resolved through a cumbersome governance structure. The G20 should step in to address this weakness. 

What is GEMs?

Established in 2009 by the European Investment Bank (EIB) and the International Finance Corporation (IFC), the GEMs Risk Database summarizes data on member organizations’ default and recovery rates by geography, sector, and time period for private, sovereign, sub-sovereign, and sovereign guaranteed lending. Access to this information enables member organizations to make more informed decisions about projects and program risks, which is especially valuable for markets where such data are scarce. 

Since 2009, GEMs has expanded to become a consortium of 24 MDBs and development finance institutions (DFIs) administered by the EIB in Luxembourg. GEMs is overseen by a steering committee composed of risk officers representing seven of its largest members (the EIB, World Bank, IFC, African Development Bank, Asian Development Bank, Inter-American Development Bank, and European Bank for Reconstruction and Development). 

Private investors have expressed interest in using this data to help them better assess risk in emerging markets, but GEMs remains available only to participating organizations due to a combination of technical challenges, legal issues, and confidentiality concerns. While admittedly a pretty prosaic issue even for the staid world of development finance, this barrier to access matters because data provision could unlock significant private investment. As explained in this CGD blog: “more transparency about the MDB track record in and of itself would be a powerful force for making and expanding markets. It would very likely challenge some preconceptions about risk, and it would help more accurately assess risk where the absence of information results in excess risk premia.”

Reforming GEMs: fits and starts

In response to external pressure from both shareholders and potential investors, GEMs issued its first public report in 2021, summarizing data on sovereign and sovereign-guaranteed default rates of 11 MDBs and DFIs and a second report in 2022 on non-sovereign data. Unfortunately, the information provided was limited: default data were not disaggregated by country or sector and no recovery rates were provided. Both are scheduled for an updated release in 2023.

There were no obvious channels for securing a more comprehensive product until mid-2022 when GEMs was catapulted from total to relative obscurity due to its prominent mention in the 2022 G20 capital adequacy framework (CAF) report. The report recommended that GEMs be transformed into a standalone entity “with strong governance, management and sustainable funding, including an independent chief operating officer supported by MDB contributing members.” The proposal garnered broad support: it was endorsed or welcomed in the 2022 Bali declaration, the chair’s summary at the World Bank 2023 Spring Meetings, the 2023 Paris declaration, the latest G20 finance ministers’ statement and the recently issued G20 report on multilateral development bank reform, to name a few. 

GEMs’ committee members themselves responded positively to the CAF recommendation and are meeting regularly to take on issues that need to be resolved for the database to become a standalone entity. However, GEMs is a consortium that operates by consensus, in good faith but with no concrete legal structure, making the decision-making process time-consuming. For example, members need to agree on how the data should be mined (e.g., by country, region, sector, income group, etc.) and committee members are struggling with how to provide data that are sufficiently granular to be meaningful while preserving confidentiality.

Another controversial issue is whether investors should be able to request—and pay for—bespoke information for a specific market or sector. As this CGD blog lays out, hard-won access to a GEMs dataset on a confidential basis enabled a Sustainable Development Goals-linked investment fund (ILX) to attract over $1 billion in financing, including from a major Dutch pension fund. This seemed a promising avenue for other funds, but GEMs members subsequently decided against offering this service because of the legal complexities it engendered.

Momentum seems to be in favor of not reviving this service, which makes it all the more important that the data offered publicly are of real value to users. GEMs members are committed to this outcome, but because other relevant parties (e.g., investors and MDB shareholders) are not involved in the process, there is a risk that the GEMs 2.0 proposal, once finalized, may disappoint.

Absent a fee-for-service option, there also needs to be clarity around a funding mechanism for GEMs. Currently, members jointly support GEMs and pay dues equally. A stand-alone entity will be more costly—on the order of $3 million or more per year, or at least $125,000 per member. This is manageable for the large MDBs but could be a significant burden on the small DFIs. This suggests that a proportional fee structure may be more appropriate, though that too could prove hard to negotiate.

What now?

The question of how to resolve these and other relevant issues is paramount. What could precipitate a faster and better decision-making process is for shareholders to adopt a more muscular role. Repeated entreaties via communiques and other public documents are simply not adequate to the task. 

The most recent G20 finance ministers’ statement has this to say: “We appreciate the ongoing collaboration among MDBs on the timely release of Global Emerging Markets (GEMs) data and the launch of GEMs 2.0 as a stand-alone entity by early 2024.” But steering committee members view this timing as unrealistic and find the time pressure to be counterproductive. Be that as it may, there needs to be some mechanism for tracking, encouraging, and assessing progress.

One option would be to provide more prescriptive language in the next G20 communique (which will be issued in September) that creates a more transparent and inclusive process that enables external stakeholders—especially private investors—to weigh in. The G20 should also elaborate on their expectations for GEMs 2.0 in terms of data provision. A collaborative effort to create a realistic timetable with clear milestones and outcomes would align expectations across stakeholders, including GEMs members, and ultimately lead to a more valuable and higher quality product.