Volatility Transfers between Cycles : A Theory of Why the "Great Moderation" Was More Mirage than Moderation

In this paper we use a New Keynesian model to explain why volatility transfer from high frequency to low frequency cycles can and did occur during the period commonly referred to as the "great moderation". The model suggests that an increase in inflation aversion and/or a reduction to a commitment to output stabilization could have caused this volatility transfer. Together, the empirical and theoretical sections of the paper show that the "great moderation" may have been mostly an illusion, in that lower frequency cycles can be expected to be more volatile, given that there has been no apparent reversal in any of the policy parameters and hence in the volatility found in the low frequency cycles identified by use of time-frequency empirical techniques. In fact, those cycles appear to have increased in power and volatility in both relative and absolute terms. Keywords: New Keynesian model, business cycles, growth cycles, time-frequency domain, discrete wavelet analysis, Empirical Mode Decomposition