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

horizons for up to two years and values around 0.5 to 1.7 for the BIG specifi-
cation, due to the additional posterior coefficient uncertainty. The results are
quite a bit higher for the Eichenbaum-Evans recursive identification: there,
the Sharpe ratio reaches values near 4 before slowly moving back to zero.
We do not view the recursive identification as plausible for this updated data
set, however, as we have argued before² . We therefore conclude, that there is
indeed a sizeable forward discount bias puzzle, offering rewards to risk which
exceed the corresponding annual US stock market Sharpe ratio by a factor
of up to 5, but that the reward for risk is not quite as extreme as suggested
by the recursive identification by Eichenbaum and Evans.

We similarly investigate the forward premium ξ_k for one-month holding
periods in figure 11 and the corresponding Sharpe ratios in figure 12. The
numbers are very similar, and thus do not change the insights.

We conclude from this that the reward for the risk of betting on violations
of the uncovered interest parity is higher by a factor but not by an order of
magnitude, compared to Sharpe ratios typically calculated for asset markets.
It may be puzzling why financial markets offer such a high reward for risk in
general. While the market for foreign exchange offers even higher rewards,
it is not drastically different from other asset markets in that respect.

The last line of figure 4 concerns the variance decomposition of the move-
ments of exchange rates: we shall discuss this together with the variance
contributed by foreign monetary policy shocks in subsection 4.3 .

² Nonetheless one may wonder, which types of shock are the cause of these reactions.
There must be some shocks which generate these kinds of high Sharpe ratios. Here we
only argue, that they are not shocks to monetary policy. If they are due to e.g. restric-
tions of capital movements or changes in taxation, these high Sharpe ratios may not be
“exploitable” for smart investors. Investigating this issue further is surely interesting but
beyond the scope of this paper.

19

<<   19   20   21   22   23   24   25   >>

More intriguing information