Diversification benefits of green bonds in China: a dynamic robust optimization approach

This paper investigates whether China’s green bonds should be included in investors’ portfolios. We employ the dynamic robust optimization approach to examine the diversification benefits derived from investing in green bonds and compare them with those derived from conventional bonds. The dynamic robust optimization method is based on minimizing the worst-case Conditional Value at Risk (WCVaR) under a mixture of dynamic R-vine copulas. We find that the dynamic robust model performs well in terms of risk-adjusted returns, and green bonds provide more diversification benefits than conventional bonds in the out-of-sample setting. The diversification benefits of green bonds are stronger over the post-2016 period due to the top-down policy support from the Chinese government. While for the green portfolios, the portfolios with the greenest green bond indices have the least diversification benefits in most cases. Overall, our findings suggest that China’s green bonds should be included in optimal portfolios but with caution.