New Evidence on the Puzzles. Results from Agnostic Identification on Monetary Policy and Exchange Rates.

shock and the concurrent on-impact movement of the exchange rate. The
investment strategy examined here relies on predictable movements, and not
on reacting more quickly than the foreign exchange market to news about
monetary policy. Put differently, as long as an investor starts this investment
strategy within the “impact month” of the monetary policy shock, he will
receive ρ_k .

(Conditional) uncovered interest parity in the context of our analysis
says that one should not be able to make (or loose) money via these hedging
strategies, i.e. ρk ≡ 0 for all k or ξj ≡ 0 for all j . This appears to be in
conflict with the evidence found in the literature, see figure 2.

The forward discount puzzle is rarely stated in these terms, though, (with
Faust and Rogers (2003), being one notable exception. Rather, the forward
discount puzzle is an implication of the observations on delayed overshooting.
Let pk and p_k be the impulse responses of the log price levels in the foreign
country resp. in the US, and let

qk = sk + (pk ^- Pk)                           (1)

be the impulse response of the real exchange rate. The Dornbusch (1976)
overshooting model results from adding to (conditional) UIP the assump-
tions, that the impulse response for the long run real exchange rate qk will
converge to zero, q_k → 0 due to long run purchasing power parity, that prices
are sticky and that the differences in the short rates i_j∙ — i* slowly reverts to
zero following the initial monetary policy shock characterized by a liquidity
effect i₀ — i0 > 0. I.e., under these assumptions, q₀ should be large and
negative and slowly revert back to zero and the domestic currency should
appreciate on impact. However, the empirical literature has found delayed
overshooting in response to US monetary policy contraction, see figure 1 and
no significant foreign-currency appreciation or even foreign-currency depre-
ciation in response to a foreign monetary policy contraction, see 3.

<<   8   9   10   11   12   13   14   >>

More intriguing information