1.
Introduction
Inflation uncertainty, its linkages with actual inflation and its potential impact on real
economic activity have been extensively analysed in the literature. Friedman (1977)
was the first to suggest that higher average inflation could result in higher inflation
uncertainty. This idea was developed by Ball (1992) in the context of a model in
which higher inflation leads to increasing uncertainty over the monetary policy stance.
The possibility of a negative effect of inflation on its uncertainty was then considered
by Pourgerami and Maskus (1987), who pointed out that in an environment of
accelerating inflation agents may invest more resources in inflation forecasting, thus
reducing uncertainty (see also Ungar and Zilberfarb, 1993). Causality in the opposite
direction, namely from inflation uncertainty to inflation, is instead a property of
models based on the Barro-Gordon setup, such as the one due to Cukierman and
Meltzer (1986).
Concerning the relationship between inflation uncertainty and real economic activity,
some authors suggest that the former reduces the rate of investment by hindering
long-term contracts (see, e.g., Fischer and Modigliani, 1978), or by increasing the
option value of delaying an irreversible investment (see, e.g., Pindyck, 1991). Others
argue that, to the extent that it is associated with increased relative price variation, it
reduces the allocational efficiency of the price system (see Friedman, 1977). In
contrast, Dotsey and Sarte (2000) show that inflation variability may increase
investment through its impact on precautionary savings. Finally, Cecchetti (1993)
suggests that a general equilibrium, representative agent model is not likely to yield a
convincingly unambiguous result on the impact of uncertainty on real economic
activity.
On the empirical side, a number of studies have investigated the relationship between
inflation and inflation uncertainty, typically adopting an econometric framework of
the GARCH type (see Engle, 1982), and providing mixed evidence (see Davis and
Kanago, 2000 for a survey, and Baillie et al., 1996, Brunner and Hess 1993,
Kontonikas 2004, Grier and Perry, 2000 for some specific contributions). Other
authors take instead a VAR approach to analyse US data. In particular, Benati and
Surico (2008) estimate structural VARs with time-varying parameters and stochastic
volatility and report a decline in inflation predictability, showing that this can be
caused by tough anti-inflation policies in the context of a sticky price model. Cogley
et al. (2009) take a similar approach but focus instead on the inflation gap, and
present evidence that this started decreasing in the early 1980s, when the inflation rate
was reduced, and kept falling when Greenspan took over the chairmanship of the Fed.
This result is then rationalised in terms of a new Keynesian DSGE model, within
which a more stable long-run inflation target is the main factor leading to lower
inflation volatility and persistence.
Empirical studies on the linkages between inflation uncertainty and real economic
activity also report conflicting results both in terms of the sign (see, e.g., Holland,
1993) and of the magnitude and timing of the effects (see, e.g., Davis and Kanago,
1996, Cunningham, Tang, and Vilasuso, 1997, and Grier and Perry, 2000). Elder
(2004) finds that in the US inflation uncertainty has significantly reduced real
economic activity. This holds for the period prior to 1979, after 1982, and over the