Sovereign loan portfolios of Multilateral Development Banks (MDBs) typically comprise a small number of borrowers, making them particularly exposed to single name concentration (SNC) risk. Using realistic MDB portfolios constructed from publicly available data, this paper quantifies SNC risk through accurate Monte Carlo simulations. We find that SNC risk can account for up to 82% of total unexpected loss in MDB sovereign loan portfolios. Comparing the exact adjustment for SNC risk with its analytical approximation currently applied by a major rating agency, we show that the approximation can overestimate SNC risk by up to 266%. This overestimation has implications for the assessment of MDB capital adequacy, potentially limiting MDB lending capacity. Our results suggest that adopting a more accurate assessment of SNC risk could increase lending capacity by approximately 5% without affecting risk-weighted capital ratios. These findings highlight the importance of refining capital adequacy methodologies to better reflect the unique risk profiles of MDBs and unlock additional lending headroom for global development. • Rating methodologies constrain lending by multilateral development banks (MDBs). • Single name concentration (SNC) accounts for up to 82% of total unexpected loss in MDB portfolios. • Approach used by rating agencies overstates SNC in MDBs’ sovereign loan portfolios by up to 37%pt. • Preferred creditor status lowers absolute, but not relative level of SNC risk. • Accurate SNC risk metrics could boost MDB development-related lending by about 5%.

For decades, gold has been considered as the most safe haven of assets in times of markets turmoil. However, this role has appears to have waned in recent years. This study aims to assess gold market behaviour over the past 37 years and examine the role of gold (and other precious metals) during periods of equity and bond market stress. Notably, we investigate whether gold lost its appeal as a safe haven asset due to its own market instability. Change point analysis, rolling mean, GARCH and DCC-GARCH approaches demonstrate that the gold market exhibits two distinct periods characterised by differing market movements, with a stable era followed by an unstable (highly volatile) era. Gold plays an insignificant role during the latter unstable period. Moreover, it exhibits a positive correlation in most high volatility periods with the S&P 500, again, especially during this latter period. This implies that gold is losing (or lost) its safe haven role during market stress and as the gold market encounters higher volatility periods, we anticipate a more positive correlation with the stock market in times of extreme stock market conditions. The outcomes challenge the prevailing definition of a gold safe haven, question the assumption of the stabilising role gold may offer to mitigate losses, and have important implications for investors seeking shelter in times of market stress. • The gold market exhibited two distinct market movements: a stable period (1980–2005) and a volatile period (2006–2024). • Gold's hedge effectiveness declines significantly during the volatile period. • DCC-GARCH and ICSS results show gold's correlation with stocks turned positive in crises post-2005, undermining its safe haven role. • Gold increasingly behaves like a risky asset, co-moving with S&P 500 volatility during turbulent periods. • Gold's safe haven role is fading in periods of high volatility.