Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Stephanie Schmitt-Groh´ e and Mart ´ ın Uribe Columbia University November 4, 2020
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Motivation • Most existing empirical work assumes that monetary policy shocks come in only one flavor, namely, in the transitory one. • However, recent work on the Neo-Fisher effect has shown that it is important to distinguish between transitory and permanent mon- etary disturbances. Permanent monetary policy shocks have been shown to be at least as important as transitory ones in explaining the dynamics of changes in output, inflation, and interest rates in the United States and Japan (Uribe, 2018) and the U.K. France, and the Euro Area (Azevedo, Ritto, Teles, 2019). • Motivated by these findings, the present paper studies the effects of monetary policy shocks on exchange rates and uncovered interest rate differentials in frameworks that distinguish between transitory and permanent monetary shocks. 2
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Related Literature I: Monetary Policy Shocks, Ex- change rates, and Uncovered Interest Rate Differ- entials • A monetary tightening causes: – the nominal exchange rate to appreciate – deviations from uncovered interest rate parity in favor of the high interest rate currency See, for example, Eichenbaum and Evans, 1995; Kim and Roubini, 2000; Faust and Rogers, 2003; Faust Rogers, Swanson, and Wright, 2003; Scholl and Uhlig, 2008; Bjørnland, 2009; Kim, Moon, and Velasco, 2017; Hettig, M¨ uller, and Wolf, 2018; Zhang, 2020. Main difference across these papers is the identification of the mon- etary policy shock (SVAR, narrative, sign restrictions, high fre- quency). 3
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Related Literature II: Permanent Monetary Shocks • Uribe (2018) and Azevedo, Ritto, Teles (2019): In a closed economy a permanent monetary tightening causes – an increase in inflation already in the short run – no output loss • De Michelis and Iacoviello, 2016: An increase in the U.S. inflation target causes – a temporary real depreciation of the U.S. dollar 4
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Main findings: • A permanent monetary tightening causes: — the nominal exchange rate to depreciate not only in the long run but already in the short run — deviations from uncovered interest rate parity against the high interest rate currency • A temporary monetary tightening causes: — the nominal exchange rate to appreciate gradually — deviations from uncovered interest rate parity in favor of the high interest rate currency • Permanent monetary shocks explain a larger fraction of the vari- ance of exchange rates and uncovered interest rate differentials than temporary ones. 5
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Permanent and Transitory Monetary Policy Shocks in an Open Economy New Keynesian Model Before going to the empirical analysis, let’s look at the predictions of an open economy NK model ` a la Gal ´ ı-Monacelli but with: – permanent monetary policy shocks and – deviations from uncovered interest rate parity. Assume incomplete international financial markets. Deviations from UIP arise due to portfolio adjustment costs (PAC) for foreign bonds as in Schmitt-Grohe and Uribe (2003). Yakhin (2020) shows that, up to a first-order approximation, a model with PAC is isomorphic to the Gabaix and Maggiori (2015) or the Fanelli and Straub (2019) model of deviations from UIP. 6
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Elements of the open economy NK model: � � 1 1 − 1 1 − 1 1 − 1 1 1 η η η – 2 goods, home and foreign: C t = (1 − v ) η C H,t + v η C F,t – Calvo price stickiness – Incomplete markets, only nominal bonds, domestic ( D t ) and for- eign ( D ∗ t ). Foreign bonds are subject to convex portfolio adjustment costs (PAC), ψ ( D ∗ t ). Budget constraint of the household: � D ∗ � P t C t +(1+ i t − 1 ) D t − 1 + E t (1+ i ∗ t − 1 ) D ∗ t − ψ ( D ∗ t − 1 = P H,t Y H,t + D t + E t t ) – Transitory ( z m t ) and permanent ( X m t ) monetary policy shocks: Y H,t + z m t + (1 − α π ) X m i t = r + α π π H,t + α y ˆ t Let uid t ≡ i t − i ∗ t − E t ǫ t +1 PAC model implies that (up to first-order) D ∗ uid t = Ψ � � t ; with Ψ > 0 7
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Impulse Responses to Permanent ( X m t ) and Transitory ( z m t ) Mone- tary Shocks in an Open Economy NK Model with PAC 1 4 0.2 2 0.1 0.5 0 0 0 -0.1 -2 0 2 4 0 2 4 0 2 4 0.5 1 0.5 0.5 0 0 0 -0.5 -0.5 -0.5 -1 0 2 4 0 2 4 0 2 4 X m z m shock shocks t t 8
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Empirical Model 9
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Variables of Interest The empirical model is an open-economy extension of Uribe (2018). Let y t log of real US output π t US inflation i t US nominal interest rate = ǫ t change in dollar exchange rate i ∗ foreign nominal interest rate t π ∗ foreign inflation t 10
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Cyclical Components of Variables of Interest The variables of interest are nonstationary. Their cyclical components are stationary yet unobservable . ˆ y t − X t y t π t − X m ˆ π t t i t − X m ˆ i t t ≡ + X m ∗ ǫ t − (1 − α ) X m ˆ ǫ t t t i ∗ i ∗ − X m ∗ t − αX m ˆ t t t π ∗ π ∗ − X m ∗ t − αX m ˆ t t t X t = permanent nonmonetary shock (output trend) X m = domestic (U.S.) permanent monetary shock t X m ∗ = foreign permanent monetary shock t 11
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks The cyclical unobservable components follow an AR process ∆ X m ν 1 ∆ X m ∆ X m t − 1 t t t y t − 1 ˆ y t ˆ z m ν 2 z m z m t − 1 t t t π t − 1 ˆ ˆ π t ν 3 ∆ X t − 1 ∆ X t ∆ X t ˆ t ˆ i t − 1 i t ν 4 z t − 1 = B ( L ) + C z t ; z t = ρ + ψ t ˆ ǫ t − 1 ˆ ǫ t ∆ X m ∗ ∆ X m ∗ ∆ X m ∗ ν 5 i ∗ i ∗ ˆ t t t − 1 ˆ t t − 1 z ∗ z ∗ t z ∗ ν 6 π ∗ π ∗ t t t − 1 ˆ ˆ t w ∗ w ∗ t t − 1 w ∗ ν 7 t t t − 1 t X m = permanent monetary shock t z m = transitory monetary shock t X t = permanent nonmonetary shock z t = transitory nonmonetary shock X m ∗ = foreign permanent monetary shock t z ∗ t = transitory foreign shock w ∗ t = UIP shock Innovations ν t ∼ iid N (0 , I ), ρ and ψ are diagonal matrices. (To simplify the exposition constants are omitted.) 12
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks 6 Observables Sample: 1974Q1 to 2018Q1. (1) ∆ y t , output growth rate. (2) r t ≡ i t − π t , interest-rate-inflation differential. (3) ∆ i t , time difference of domestic nominal rate. (4) ∆ ǫ t , time difference of devaluation rate. (5) ∆ i ∗ t , time difference of foreign nominal rate. (6) ǫ r t , real devaluation rate. Foreign country either United Kingdom, or Japan, or Canada. 13
Schmitt-Groh´ e and Uribe Exchange Rates and Uncovered Interest Differentials: The Role of Permanent Monetary Shocks Observation Equations We can then link the unobservables to the observables as follows: ∆ y t = y t − ˆ ˆ y t − 1 + ∆ X t = ˆ r t i t − ˆ π t i t − 1 + ∆ X m = ˆ i t − ˆ ∆ i t (1) t ǫ t − 1 + (1 − α )∆ X m − ∆ X m ∗ ǫ t − ˆ ∆ ǫ t = ˆ t t ∆ i ∗ i ∗ i ∗ t − 1 + ∆ X m ∗ + α ∆ X m = ˆ t − ˆ t t t ǫ r π ∗ t − ˆ = ˆ ǫ t + ˆ π t t 14
Recommend
More recommend
