This article examines Taiwan data to determine whether or not output asymme
trically responds to monetary policy shocks. Two asymmetries are defined. F
or the first asymmetric effect, referred to as the TE effect, an inverted L
-shaped aggregate supply curve with negative-sloped equilibrium locus is pr
oposed. This AS curve implies that the effect of an easy monetary policy di
ffers in different inflation regimes. The easy monetary policy is expected
to have a positive effect. no effect, and a negative effect on output durin
g low, high. and very high inflation regimes, respectively. Our results sup
port this hypothesis. Results regarding the second asymmetric effect, refer
red to as RE asymmetry, are contradictory. Different classifications of rec
essions yield contradictory results. By employing official classification,
monetary policy shocks support the RE asymmetric hypothesis. The data, howe
ver, are not overwhelmingly in favor of the RE asymmetric hypothesis using
Markov switching dating. Because the dates of recessions are markedly diffe
rent for the two classifications, the results demand a more in-depth probe
into business cycle dating. (C) 2000 Society for Policy Modeling. Published
by Elsevier Science Inc.