Page 128 - Special Topic Session (STS) - Volume 4

STS571 Jaanus Kroon

Mobile phone and credit card data: Experience
from 10 years of public private partnership
Jaanus Kroon
Eesti Pank / Bank of Estonia, Tallinn, Estonia

Abstract
Since 2008, Eesti Pank, the central bank of Estonia, has been cooperating with
private companies to source big data for official statistics. The central bank
uses mobile phone data from Estonian mobile network operators to quantify
inbound and outbound travel, and credit card payment data to calibrate
expenditure figures. The combination of these two datasets and several other
sets of reference data allows timely and efficient production of trade in
services statistics, used in the compilation of the balance of payments and the
national accounts. The cooperation has stood the test of time as part of the
statistical business process at the central bank. After a full decade of
cooperation between official statistics and big data providers in Estonia, we
look back at the experience gained and the lessons learned:
• the evolution of mobile phone data and credit card data over the years
• the ways both types of data can be used for official statistics
• lessons from the partnership
This best practice can be used by statisticians in accessing and adopting data
sources to produce official statistics in their own countries.

Keywords
big data; mobile positioning; central bank statistics; balance of payments;
cooperation for official statistics.

1. Introduction
Globalisation, the blurring of borders and the complexity of measuring
cross-border transactions have posed a considerable challenge for external
statisticians for some time now. The rapid growth in worldwide travel,
membership of the Schengen Area, where there are no border controls and so
no data collection at borders, and the discontinuation of regular border
surveys by Statistics Estonia because of budget cuts forced Eesti Pank as the
institution responsible for external sector statistics to find an alternative way
to continue the border-crossing time series. Border-crossing data are an
important input in the compilation of the country’s monthly and quarterly
balance of payments, where exports and imports of travel services play a major
role. Many alternative options were explored to meet the demand for a high
quality and efficient data source at a reasonable cost and with low labour

117 | I S I W S C 2 0 1 9