台灣需要總體審慎政策嗎？如何做？與貨幣政策如何搭配？

Does Taiwan Need a Macro-Prudential Policy? How to Do It? How to Coordinate with Monetary Policy?

本文建構了一個包含擔保品借貸限制的小型開放經濟體動態隨機一般均衡（dynamic stochastic general equilibrium, DSGE）模型，透過貝式估計法（Bayesian estimation），將其應用在台灣的經濟環境，並運用動態隨機一般均衡暨向量自我迴歸（DSGE-VAR）架構來評估模型的配適度。我們依此模型來分析貨幣政策與總體審慎政策的福利意涵。研究結果顯示，考慮房價或國內貸款的利率政策，可以增進社會整體與個別成員的福利，而針對房價或國內貸款採取反景氣循環的貸款成數（loan-to-value）限制，亦可以達到柏拉圖改善的效果。除此之外，利率政策適切搭配貨款成數限制政策，則可以獲致較大的社會福利，並有效地維持金融穩定、降低通膨變化。然而，由於個別族群的異質性，需設計不同的政策組合，方能極大化不同族群的福利。

This paper develops a small open economy dynamic stochastic general equilibrium (DSGE) model with collateral constraints. We estimate the model with data in Taiwan using the Bayesian technique, and assess the fit of the model by the DSGE-vector autoregression (DSGE-VAR) methodology. We demonstrate the welfare implications of the monetary policy and the macro-prudential policy. Our results suggest that an interest rate rule optimally responding to the growth of housing prices or domestic loans could enhance social welfare and make all agents better off. Introducing an optimal loan-to-value (LTV) ratio rule in response to domestic credit or housing price dynamics is also Pareto improving. Furthermore, the optimal deployment of both the monetary and macro-prudential policy could yield more social welfare. The coordination of policies effectively stabilizes the financial system and limits the volatility of inflation. However, because of the heterogeneity of agents, different mixes of policies require customized design for individual welfare optimization.