Explaining the failure of the unbiased forward rate hypothesis using a time -varying risk premium
This dissertation uses a time-varying risk premium to explain the failure of the unbiased forward rate hypothesis. In our opinion the risk premium drives a wedge between the forward rate and the expected future spot rate. We derive the risk premium from a small open economy continuous-time partial-equilibrium model. An exogenous mixed jump-diffusion exchange rate process is used in order to explicitly consider the effect of foreign currency intervention on the risk premium. The ability of the representative investor to hedge his foreign exchange risk stemming from his investment in foreign assets is also recognized.^ We are concerned with answering the following questions: Is there empirical evidence supporting our hypothesis that a time-varying risk premium is responsible for the failure of the unbiased forward rate hypothesis? What effect does foreign exchange intervention have on the risk premium? What effect does forward exchange hedging have on the risk premium? Are these effects significant?^ We find that our optimal foreign asset demand equation differs from the traditional foreign asset demand equations of Frankel (1982), Branson and Henderson (1985) or even Svensson (1992). With the ability to hedge his foreign exchange risk in the forward market the representative investor no longer needs to hold foreign assets in proportion to future foreign consumption shares.^ We offer a finding of cointegration as evidence that a time-varying risk premium is, at least partly, responsible for the failure of the unbiased forward rate hypothesis. We have determined using the Engle-Granger and Johansen cointegration procedure that the current spot exchange rate, lagged forward rate and lagged risk premium are cointegrated. These results are reinforced by the results of the Saikkonen procedure. Hypothesis tests indicate that the risk premium is significant and cannot be excluded from the long-run relationship. Foreign currency interventions are found to have an insignificant impact on the spot exchange rate, the risk premium and the UFRH. ^
"Explaining the failure of the unbiased forward rate hypothesis using a time -varying risk premium"
(January 1, 1998).
ETD Collection for Fordham University.