920 Monetary Policy, Beliefs, Unemployment and Inflation
by bringing together what have hitherto been treated as two distinct – and largely
separate – possibilities for regime change: that the 1980s labor market reforms had
significant effects on unemployment, and the possibility that important effects
on inflation and unemployment followed from switching to inflation targeting
and central bank independence. The first of these two possible sources of regime
change refers to a large “shocks versus institutions” literature (see Blanchard and
Wolfers, 2000, amongst others). The other relies on the pioneering research by
Sargent (1999) which derives optimal monetary policy in a natural rate model
where expectations are formed by recursive learning. As with other research on the
policy model, section 18.5 considers to what extent a model of optimal monetary
policy with learning conforms to the broad features of UK inflation since 1980. To
do this, however, it amends the policy model in the light of econometric results on
the determinants of long-run unemployment from section 18.4. These econometric
results conclude that the evidence does not support the “wage push” interpretation
of changes in long-run unemployment. Instead, it is shown that extensions to the
unemployment model that give a major role to international factors, such as real oil
prices and measures of international competitiveness, are needed for the model to
capture the broad movements in UK unemployment. Once this extension is made
to the policy model, it is found that optimal inflation solutions from it conform
to the broad pattern of changes to inflation observed over the last 25 years or so
without appealing to any regime change. As is evident in this account, the break-
point for possible labor market regime changes is the early 1980s and monetary
policy regime change is taken to be from the end of 1992 since, as is evident
from references listed later, these are the alleged break-points in most of the labor
market and monetary policy debates in the UK. Hence, we do not attempt any sort
of estimation and inference about likely break-dates, which is an important, but
separate, topic to what is presented here.^2
18.2 A selection of background literature
18.2.1 A baseline New Keynesian policy model
It is helpful, pedagogically, to relate the principal papers referred to later to a “base-
line” New Keynesian policy model (NKPM), since they can often be viewed either
as a form of complete NKPM or as single equations from it, such as the aggregate
supply (AS) or the policy rule for interest rates (the “Taylor rule”). The aggregate
demand (AD) equation has received less attention in the literature, and that is the
line adopted here.^3 The example below is the closed-economy model found in
Henry and Pagan (2004). This baseline NKPM is given by equations (18.1)–(18.3),
and microfoundations of these equations are discussed in,inter alia, Svensson
(2000).
πt=αππt− 1 +( 1 −απ)πte+ 1 +αy(yt− 1 −y∗t− 1 )]+ξt (18.1)yt=βyyet+ 1 −σ(rt−πte+ 1 )+vt (18.2)rt= ̄rt+β(πte+ 1 −π∗)+γ(yt−y∗t). (18.3)