不对称风险下的宏观审慎政策:贝叶斯结构分位数 VAR 方法

Macro-prudential policymakers face a risk-management problem: they must assess medium-term downside risks to the real economy stemming from financial imbalances and the associated threat of future financial crises, while accounting for the cost of policy interventions in terms of foregone upside potential. This trade-off between limiting downside risks and preserving upside growth is inherently asymmetric and depends on the distribution of future macroeconomic outcomes. We formalize this decision problem by defining novel measures of downside risk and upside potential and by stipulating an explicit loss function for macro-prudential policy. To evaluate policy in this setting, we develop and apply a novel Bayesian structural quantile vector autoregressive model that delivers flexible and potentially asymmetric predictive distributions. Using quantile impulse responses, we assess the scope for macro-prudential policy to reduce growth volatility and negative skewness, infer the macro-prudential policy stance, and characterize the circumstances under which macro-prudential interventions are likely to be beneficial.