Half a Trillion in Dry Powder? The Time for MDB Financial Reform Is Now
Published: September 01, 2021
To help developing countries meet their financing needs, the multilateral development bank (MDB) system needs to get much bigger. Key to a bigger MDB system is a more financially efficient one. Instead, a conservative capital adequacy policy could be holding back nearly half a trillion in collective MDB financial firepower (the equivalent of almost the entire allocation of IMF special drawing rights). While the G20 has repeatedly advocated for MDB balance sheet optimization (see the G20 balance sheet action plan), progress has been incremental at best. This is largely because shareholders and the MDBs themselves do not necessarily agree on the optimal balance between financial ambition and prudence. But in the wake of the COVID crisis and the imperative of scaled up financing to meet the climate challenge, now is the time to reinvigorate the MDB capital efficiency agenda.
What’s the problem?
Thanks to high credit ratings (usually AAA), MDBs can access funds cheaply on the market and on-lend these to clients at significantly lower rates than those clients could otherwise access on their own. This model has proven remarkably effective. For example, the International Bank for Reconstruction and Development (IBRD), the World Bank’s hard loan lending window, has received around $18 billion in shareholder equity since its inception over 70 years ago, enabling it to provide over $700 billion in loans.
Maintenance of AAA bond ratings has led the finance departments of MDBs to prioritize safety over efficiency. This is compounded by rating agencies’ idiosyncratic and evolving methodologies for evaluating these unique institutions which have left MDBs contorting between different financial targets. The result is that MDBs often end up holding high levels of capital; funds which could otherwise be put to productive development use.
There is good reason to question whether targeting less conservative financial ratios would jeopardize the MDBs’ AAA ratings (Chris Humphrey has persuasively made the case that it would not here). MDBs have strong underlying fundamentals underpinned by their preferred creditor status, backing from their shareholders (often partly in the form of callable capital), and uniquely low loan impairment rates (when principal and interest are not repaid in full). The IBRD, for example, has a 0.2 percent nonaccrual rate (the percentage of loans over 89 days overdue), a fraction of the nonaccrual rate for commercial loans in the United States, yet holds significantly more capital than private sector peers.
As a result of the rating agency methodologies, MDB capital adequacy policies are also significantly more conservative than their charters permit: on average MDBs’ available statutory headroom is generally four times larger than their headroom based on their prudential limits (the AfDB is an outlier here).
MDB statutory lending limits are the legal limits in their charters that define how much they can lend relative to the amount of capital they have on their balance sheets—including both paid-in and callable capital (where relevant; not all MDBs have callable capital). Prudential limits are the capital adequacy policies that the MDBs themselves adopted and often have a narrower interpretation of equity (i.e., capital that has been paid-in versus callable from shareholders in case of need) than the statutory lending limits allow.
Financial efficiency diverges widely across the MDB system. Equity-to-loans (E/L) ratios are a useful metric for gauging MDB financial efficiency: they show how much development financing shareholder equity is mobilizing. With an E/L hovering around 22-23 percent, the IBRD is the most financially efficient MDB—by virtue of its size and client diversity—and each dollar of equity mobilizes around 5 dollars in lending. In contrast, the regional development banks tend to operate on a 1:3 basis, and the private sector arms closer to 1:2. IDB Invest is a relative outlier because it is a new institution still establishing its track record. Similarly, IDA, the arm of the World Bank that lends to the poorest countries, has only recently started to issue debt and has a different business model than the other institutions, and consequently has significantly more conservative ratios. (For a deeper dive into IDA financial issues, see here.)
Looking at how MDB E/Ls evolved over the course of the COVID-19 pandemic is also interesting. For the most part—with the notable exception of the Asian Development Bank (ADB)—E/Ls have remained relatively flat from 2019 to 2020. This reality contrasts with the global financial crisis experience, where MDBs showed a general willingness to relax their E/Ls. For instance, the IBRD’s E/L ratio shot down from 38 percent in 2008 to 27 percent in 2012 and the institution took the extraordinary measure of revising its E/L policy minimum to accommodate a surge in funding. ADB went from 39 to 33 percent E/L over this period and the IDB from 38 to 30 percent. All the institutions got capital increases after the crisis and, except for the IBRD, they have since returned or surpassed pre-global financial crisis E/Ls. These differences translate in the aggregate lending numbers: the MDB system provided $116 billion in funding in 2019 compared to $153 billion in 2020—a not insignificant increase of 31 percent, but far less than the 255 percent surge that took place in the aftermath of the global financial crisis. It is unclear if the lack of a financing surge today commensurate with global financial crisis levels is due to a lack of demand or financial rationing on the part of MDBs who have not received clear instructions from their shareholders and are uncertain about the prospects for future capital.
More ambitious financial targets could generate significantly more lending. The table above provides an illustrative scenario of how the MDBs’ headroom could expand if they were to target slightly lower E/L ratios while still remaining well above the standard 10 percent benchmark of a commercial bank. In order to account for MDB idiosyncrasies, we assign different indicative E/L targets to institutions based on their specific characteristics, namely client type, regional versus global reach, and availability of callable capital. At one end of the spectrum, we assign IBRD the most ambitious gearing target (15 percent) since it benefits from a strong credit profile, exposure diversification across many regions, and callable capital. At the other end of the spectrum, we assign more conservative targets for institutions that operate in riskier markets (the private sector or non-creditworthy lower-income governments), do not have callable capital and/or have concentrated exposures. While these E/L targets are conservative in comparison to those of private financial institutions like commercial banks, which do not have some of the business profile advantages of MDBs, the payoff is impressive: an additional $400 billion in exposure headroom for MDBs to put towards development.
Last April, the G20 agreed to launch an independent review of the MDB capital adequacy frameworks to assess their financing positions and capacity. To be credible, the review will need to be veritably independent and have access to each MDB’s financial data (no small ask!). And though the mechanics of MDB balance sheets may seem mired in technicality, decisions around risk appetite are ultimately political. Moving the needle will require shareholders to orchestrate a proactive and coordinated push for MDB to adopt an approach that better balances risk and financial efficiency. With large questions looming for the MDB system around its role in a rapidly changing and increasingly crisis-prone world, today more than ever, maximizing MDB financial firepower should be a central piece of the MDB modernization agenda.