Robust portfolio optimization for banking foundations: a CVaR approach for asset allocation with mandatory constraints

Abstract

This paper focuses on an innovative asset allocation strategy for risk averse investors who operate on very long-time horizons, such as endowments and the Italian foundations of banking origin (FBOs). FBOs play a pivotal role in supporting economic, financial and sustainable growth in the long term. In the search for a model which optimizes FBO portfolio choices in the light of regulatory constraints on their sizeable investable portfolio, we highlight the risk-adjusted performances obtained using a robust conditional VaR (R-CVaR) approach—assuming different risk profiles—which corrects some of the Markowitz approach pitfalls and accounts for tail risk. We compare the two models using a buy and hold strategy: the R-CVaR delivers better returns than a Markowitz portfolio, even when those performances are measured with a mean–variance metric.