Regulatory risk constraints and investment decisions: unintended consequences for the financial system.

This thesis contributes to the extant research on the impact of regulatory constraints on financial markets, by presenting a collection of three intertwined essays. The first essay examines the relation between regulatory constraints on market risk and the fluctuation of the financial market, often depicted in the literature but not empirically proven. I find that market volatility is significantly dependent on (and Granger-caused by) the relative market risk exposure of Italian banks, measured as the ratio of value at risk and banks’ market risk limit. In the second paper, I explore the channel for the results obtained in the first paper. The theoretical framework is based on a risk-constrained mean-variance framework. In such a framework, if the constraint binds, the portion of the portfolio invested in risky assets (alpha) is lower than in the unconstrained scenario, as expected. Furthermore, alpha is inversely related to the relative market risk exposure (as above, given by the ratio between value at risk and market risk limit). Empirical tests confirm that this constrained mean-variance framework is more accurate in forecasting investment behaviour in risky assets of the Italian banks than the ordinary mean-variance framework. In the third essay, I investigate the role of uncertainty, whose relevance has been deeply investigated by several papers. By adding uncertainty to the constrained mean-variance framework built in the previous chapter, I find that an increase in uncertainty determines a decrease of the portion of portfolio invested in risky assets, in line with the literature. I perform empirical tests which confirm the theoretical result, especially in high-volatility periods when constraints bind tighter.