Gunnar Bårdsen and Ragnar Nymoen 873are trying to attain a lower real wage by nominal price increases, at the same time
as the wage bargain is delivering nominal wage increases that push the real wage
upwards.
Bårdsenet al.(2005, Ch. 6) show which restrictions on the parameters of the sys-
tem (17.30) are necessary forut→uss=uwto be an implication, so that the NAIRU
corresponds to the stable steady state. In brief, the model must be restricted in such
a way that the nominal wage- and price-setting adjustment equations become two
conflicting dynamic equations for the real wage. Because of the openness of the
economy, this is not achieved by imposing dynamic homogeneity. What is required
is to purge the model (17.30) of all nominal rigidity, which seems to be unrealistic
on the basis of both macro- and micro-evidence.
We have seen that the Layard–Nickell version of the NAIRU concept corresponds
to a set of restrictions on the dynamic model of wage- and price-setting. The same is
true for the natural rate of unemployment associated with a vertical Phillips curve,
which still represents the baseline model for the analyses of monetary policy. This is
most easily seen by considering a version of (17.28) with first-order dynamics and
where we simplify the equation by setting the short-run effects of productivity,
unemployment and taxes equal to zero (β 12 =β 14 =β 15 = 0 ). With first-order
dynamics we have:
wt−α12,0qt=c 1 −γ 11 ecmbt− 1 +β 18 pt+ (^1) t,
and using (17.22) we can then write the wage equation as:
wt=kw+α12,0qt+β 18 pt−μwut− 1 (17.34)
−γ 11 (wt− 1 −qt− 1 )+γ 11 ( 1 −δ 12 )(pt− 1 −qt− 1 )+γ 11 δ 16 T (^1) t− 1 + (^1) t,
wherekw=c 1 +γ 11 mw, and the parameterμwis defined in accordance with Kolsrud
and Nymoen (1998) as:
μw=γ 11 δ 13 whenγ 11 >0orμw=φwhenγ 11 =0. (17.35)
The notation in (17.35) may seem cumbersome at first sight, but it is required to
secure internal consistency: note that if the nominal wage rate is adjusting towards
the long-run wage curve,γ 11 >0, the only logical value ofφin (17.35) is zero, since
ut− 1 is already contained in the equation, with coefficientγ 11 δ 15. Conversely, if
γ 11 =0 so the model of collective wage-bargaining fails, it is nevertheless possible
that there is a wage Phillips curve relationship consistent with the assumedI( 0 )-
ness of the rate of unemployment, henceμw=φ≥0 in this case.
Subject to the restrictionγ 11 =0, and assuming an asymptotically stable steady-
state inflation rateπ, (17.34) can be solved for the Phillips curve NAIRUuphil:
uphil=
kw
φ
- (α12,0+β 18 − 1 )
φ
π,