The Preparation of Monetary Policy pp 87-107 | Cite as

# Measuring Inflation Expectations: A Survey Data Approach

## Abstract

Expectations of inflation can be important determinants of actual inflation because they influence the process of wage bargaining, price setting, speculative buying and financial investing decisions (Englander and Stone, 1989). A central bank, given its primary objective of price stability, therefore takes a natural interest in inflationary expectations of economic agents. In fact, as illustrated by the theoretical work of, e.g., Kydland and Prescott (1977) and Barro and Gordon (1983), inflationary expectations of the public are an important factor determining the success of the central bank in achieving its policy goal. Measures of inflation expectations can be used for monetary policy purposes, for example as information variables in the sense defiined by Friedman (1993), or as intermediate targets in the so-called direct inflation targeting strategy (Svensson, 1996; Fuhrer, 1997).

## Keywords

Monetary Policy Unit Root Forecast Error Inflation Rate Vector Error Correction Model## Preview

Unable to display preview. Download preview PDF.

## References

- 2.The Carlson-Parkin or CP-method was fiirst employed by Theil (1952) in order to provide an alternative theoretical ju st i fication for Andersons’ (1951, 1952) use of ba lance stat ist ics to quantify the results obtained from the monthly surveys held by the IFO Institute in Munich. Balance statistics are defined as the difference between the percentage of respondents who report an increase of, say, prices, and the percentage of respondents reporting a decrease.Google Scholar
- 3.Both questions also include a ‘don’t know’ category. In what follows, we allocate the numbers of this category proportionally to the other response categories. See Visco (1984, pp. 30,32) for a discussion.Google Scholar
- 4.For monetary policy, the dispersion measure of this distribution is also of interest, since it proxies the inflation uncertainty expected by consumers (Juster and Taylor, 1975; Chan-Lee, 1980). Our method also generates estimates of this dispersion measure, but space limitations force us in this paper to concentrate on the information content of the centrality measure.Google Scholar
- 5.For details regarding this and the other distributions used in this paper, see Johnson and Kotz (1970).Google Scholar
- 6.Also, the maximum negative outcome is by defiinition limited, whereas the maximum positive response is, in theory, unbounded. This also suggests skewness to the right.Google Scholar
- 7.The preference of Carlson (1975) for the scaled t-distribution was based on empirical, not theoretical, grounds. In turned out that this distribution provided the best fit to the observed subjective distributions of respondents to the Livingston survey in the US.Google Scholar
- 9.The implication is that the perceived inflation rate is an unbiased estimate of the actual inflation rate. However, this does not imply that the expected future inflation rate, which is the product of the perceived rate and a nonlinear term representing the frequencies of the assumed cumulative distribution function (see the Appendix), is by construction an unbiased estimate of the inflation rate.Google Scholar
- 10.The data are published in the Statistical Bulletin of the Dutch Central Bureau of Statistics or in European Economy, Supplement B, of the European Commission of the EC.Google Scholar
- 11.For a description of Theils inequality coefficient, see Koutsoyannis (1977, pp. 492–497).Google Scholar
- 12.This point is especially important in our case because of the relatively short span of the sample period (1986:04–1997:02; 130 observations). This adversely affects the power of the unit root and the cointegration tests.Google Scholar
- 13.Such a correction is necessary given the fact that the forecast horizon (in our study 12 months) exceeds the sampling interval (1 month in our case) (See Brown and Maital, 1981; Papadia, 1983). OLS in this case generates consistent parameter estimates, but inconsistent estimates of the covariance matrix, because, in violation of the classic OLS-assumptions, the disturbances are not serially uncorrelated but follow an MA(11) process. See Hansen and Hodrick (1980), Hansen (1982) and Hamilton (1994) for details.The Newey-West standard errors are asymptotically consistent in the presence of autocorrelation as well as heteroskedasticity.Google Scholar
- 14.Based on a visual inspection of the data, we decided against assuming deterministic trends in the data when constructing the VAR on which the likelihood ratio tests are based. The VAR included 12 lags.Google Scholar
- 15.In addition to unbiasedness, other necessary conditions for rationality include efficiency and orthogonality. Both conditions imply forecast errors that are, in general, not serially correlated. Empirical studies by and large reject the rationality of inflation expectations measures; see, for example, Batchelor and Dua (1987), Evans and Gulamani (1984), Holden and Peel (1977), Pesando (1975), De Menil and Bhalla (1975), De Leeuw and McKelvey (1981), Madsen (1996), Pearce (1979), Pesaran (1985), Thomas (1995) and Figlewski and Wachtel (1981).Google Scholar
- 16.i.e. additional to the traditional Granger causality tests. Note that the latter should be formulated in terms of changes in (expected) inflation because of the persistence.Google Scholar
- 17.In a relatively closed economy such as the euro-area, the benefit of exchange rate stability vis-a-vis a major trading partner (such as the US) would not outweigh the costs of losing one’s ability to tailor monetary policy specifically to domestic economic conditions. Given aGoogle Scholar
- 18.For a recent discussion of this topic, see King (1996), Green (1996), Bernanke and Mishkin (1997).Google Scholar
- 19.These are the basic concepts of stability and predictability for evaluating intormation variables; see Shigehara (1996) for details.Google Scholar
- 20.For a related discussion on the mechanics of implementing inflation targeting in practice, see Baumgartner, Ramaswamy and Zettergren (1997).Google Scholar
- 21.See Figlewski and Wachtel (1981) on the effects of differences of expectations between individuals.Google Scholar
- 22.These authors estimate time varying thresholds within a traditional CP-framework with a model of stochastic parameter variation proposed by Cooley and Prescott (1976).Google Scholar
- 23.Only four of the five possible response categories are independent because the total of the proporties must equal 1.Google Scholar