本文藉由馬可夫轉換模型（Markov Regime-Switching Model），探討物價變動條件下，貨幣政策施行是否對美國不動產投資信託市場（Real Estate Investment Trusts, REITs）存在不對稱性效果，樣本期間選定1972年至2010年之月資料。先在現金流量折現模型的基礎下，建立總體經濟變數與REITs報酬率之相關理論架構，再利用自我迴歸整合移動平均模型（Autoregressive Integrated Moving Average Model, ARIMA）試圖區分預期與非預期的貨幣政策，並進一步應用門檻自我迴歸模型（Threshold Autoregressive Model, TAR）與移動平均模型（Moving average, MA）捕捉高低物價成長效果。本文採用貨幣供給（M2）成長率與聯邦資金利率變動量兩種不同貨幣政策代理變數進行頑強性檢測，實證結果顯示：在考慮高低物價成長期間下，非預期的貨幣政策衝擊對EREITs報酬率與非預期的聯邦資金利率變動對MREITs報酬率呈現不對稱性效果，其中非預期的寬鬆貨幣政策衝擊對EREITs報酬率與非預期的聯邦資金利率調降對MREITs報酬率的影響亦呈現不對稱性。此外，觀察非預期的貨幣政策衝擊與進一步觀察非預期的寬鬆貨幣政策衝擊對REITs報酬率在高低物價成長期間的效果變化，物價成長差異與貨幣政策之抵換關係（trade off）存在於REITs市場，此現象解釋為何REITs報酬具有抵抗物價上漲效果，尤其在高物價成長期間此抵換關係會較低物價成長期間更劇。 This study investigates whether the effects of monetary policy exist in the U.S. real estate investment trusts (REITs) market utilizing Markov regime-switching model over the period January 1972 to December 2010. We establish the theoretical framework by the discounted cash flow model to examine the effects of macroeconomic variables on REITs returns, and further to separate into the expected and unexpected monetary policy based on ARIMA model. Then we apply TAR model and MA model to capture the realized effects of change in price level. For the robustness we use money aggregates (M2) and Federal Funds Rate (FFR) as proxies for the stance of monetary policy. The finding indicates that unexpected monetary policy has the asymmetric effects on EREITs returns in high/low price growth period, while the asymmetric effects on MREITs returns are only found in the unexpected change of FFR. Furthermore, the effects of unexpected changes in expansionary monetary policy on EREITs returns and the effects of unexpected reductions in FFR on MREITs returns exhibit asymmetries in high/low price growth period. The findings have an implication for the effects of trade-off relationships between price growth and monetary policy exist in REITs sector. It will explain why REITs returns have inflation-hedging function. In particular, the effects in the high price growth period will be larger than in the low price growth period.
