Page 129 - Invited Paper Session (IPS) - Volume 2

IPS188 Bruno Tissot
census-type surveys to collect the prices of a list of products in selected retail
points and aggregated with proper weights to reflect the composition of
households’ consumption basket.

Nevertheless, important limitations still hinder a timely and

comprehensive provision of reliable inflation indicators across the globe. Five
key issues are worth highlighting from this perspective.

First, a robust statistical infrastructure is required to produce inflation data
on a regular basis; this calls for sufficient staff resources, adequate statistical
skills, effective IT support, the set-up of specific processes to capture prices
observed in various market segments, etc.
Second, the impact of innovation and digitalisation is posing practical
and theoretical difficulties as regards the definition and the measurement of
inflation (see Reinsdorf and P Schreyer (2019)). In particular, it is increasingly
difficult to capture instant prices offered during limited periods of time and/or
for specific groups of buyers. Moreover, correctly measuring inflation requires
the ability to capture, and correct for, quality changes. Yet, increased
innovation has reduced the ability to observe the repeated sales of the same
products, since their characteristics are constantly evolving, hence making
difficult the observation of prices for constant quality goods. One solution is
2
to apply the so-called hedonic method. But this requires capturing a wealth
of product characteristics that have to be analysed using econometric
techniques – making the data collection and compilation process more
complex.
Third, inflation is a multiform phenomenon that varies across sectors
and economic agents, leading to a multiplicity of inflation indicators. In
general, one will refer to inflation as the evolution in the prices of goods and

Defined as a “regression technique used to estimate the prices of qualities or models that are
2
not available on the market in particular periods, but whose prices in those periods are needed
in order to be able to construct price relatives”; cf OECD Glossary of Statistical Term, available
at stats.oecd.org/glossary/index.htm.
116 | I S I W S C 2 0 1 9