1 Introduction
In this paper I ask whether discretionary monetary policy can dominate policy designed ac-
cording to the timeless perspective and answer in the affirmative. I then examine the factors
that govern this result, employing a microfounded dynamic stochastic general equilibrium
(DSGE) model to ascertain the role that nominal and real rigidities play in determining
whether discretion is superior. Indeed, I show that discretion is more likely to dominate
timeless perspective policymaking in models where nominal and real rigidities are important.
Two additional contributions of the paper are that it develops a measure of conditional loss
suitable for consistently evaluating timeless perspective and discretionary policies and that it
shows how timeless perspective equilibria can be obtained from the solution to an unmodified
formulation of the optimal commitment problem. An important conclusion of the paper is
that studies applying the timeless perspective might also usefully compare its performance to
discretion.
The timeless perspective approach to policy design was first outlined in Woodford (1999a),
advanced as a solution to the “initial period” problem that characterizes optimal commitment
policies.1 At that time, Woodford (1999a) argued that the initial period problem could be
overcome if the central bank were to “adopt, not the pattern of behavior from now on that
would be optimal to choose, taking expectations as given, but rather the pattern of behavior
to which it would have wished to commit itself to at a date far in the past, contingent upon
the random events that have occurred in the meantime.” Simply put, the initial period
problem ceases to be a problem when the initial period has long since passed.2 In subsequent
work, the concepts of timeless perspective policymaking and timeless perspective equilibria
have been refined and made formal.3 Because this approach overcomes the initial period
problem, the literature on monetary policy has embraced it, to the point where such policies
increasingly form the backbone of policy analysis, and one central bank—Norges Bank—
employs the timeless perspective to construct its public interest rate forecasts.
Timeless perspective policies are closely related to optimal commitment policies. In partic-
ular, both policies involve auxiliary state variables whose role is to track the value of commit-
1As explained in Section 2, optimal commitment policies are subject to an “initial period” problem because
such policies involve the central bank exploiting expectations held as of the optimization date while promising
never to do so in the future.
2Notice that for this to be true the timeless perspective equilibrium must be stationary, a condition stronger
than standard transversality conditions. In a linear-quadratic model, the standard transversality condition
requires the economy to grow at a rate no greater than -^=, where β ∈ (0,1) is the discount factor.
3See Woodford (2003), Giannoni and Woodford (2002a,b), and Benigno and Woodford (2003, 2006).