Page 128 - Invited Paper Session (IPS) - Volume 2

IPS188 Bruno Tissot

The role of big data and surveys in measuring
and predicting inflation
Bruno Tissot
Head of Statistics and Research Support, Bank for International Settlements (BIS), Basel,
Switzerland, and Head of the Secretariat of the Irving Fisher Committee on Central Bank
Statistics (IFC)
1

Abstract
Monitoring and forecasting short-term price developments is an important
issue for public authorities. Measuring inflation requires significant resources,
and substantial methodological work has been developed over the past
decades to support this task. The recent emergence of Big Data can provide
many opportunities in this context, for instance to produce timelier indicators,
enhance the collection of specific types of prices, take due consideration of
economic agents’ expectations, and facilitate shortterm forecasting. However,
there are important challenges when using big data-type of information, and
traditional statistical surveys have continued to prove their usefulness. This
suggests that measuring and forecasting inflation should continue to benefit
from drawing on multiple, complementary approaches.

Keywords
Prices; Public policy; Internet; Inflation expectations; Nowcasting

1. Introduction
Measuring and predicting the evolution of prices is key in a market
economy, and it has therefore been traditionally an important objective for
authorities. Indeed, past centuries were characterised by various episodes of
strong money creation, leading to peaks in inflation with often important
social and political consequences (see Graph 1 for an historical perspective
on European inflation). Episodes of price deflation have been also particularly
disruptive, and the Great Depression of the 1930s underlined the need for
properly measuring and anticipating both the real economic activity and
the evolution of prices. This was clearly a key factor driving the subsequent
development of the Systems of the National Accounts (SNA; European
Commission et al (2009)) framework. Today, most advanced economies and
a growing number of developing economies have reliable statistics to
measure consumer price inflation (CPI) based on detailed international
guidance (ILO et al (2004)). Such indicators are often based on statistical,

1 Assistance from Anamaria Illes and Antonio Perrella with the preparation of the graphs is
gratefully acknowledged.
115 | I S I W S C 2 0 1 9