Page 365 - Contributed Paper Session (CPS) - Volume 2

CPS1888 Chen C. et al.

Transforming stats NZ’s quarterly financial
statistics using administrative data
Chen Chen, Mathew Page
Stats NZ, Christchurch, New Zealand

Abstract
Stats NZ has moved to an ‘administrative data first’ paradigm in the
production of economic statistics. This was driven by the increased demand
for business data and the pressure to reduce respondent burden and create
statistical efficiency. In this paper we introduce a new statistical design, which
maximises the use of administrative data in producing business statistical
outputs. We outline a model assessing where and how the New Zealand
Goods and Services Tax (GST) administrative data can be used as the primary
data source for quarterly financial statistics, as well as discuss the statistical
challenges faced in making greater use of GST data. We also present
applications with results that allow us to expand the design to a wider range
of economic outputs.

Keywords
administrative data; financial statistics; statistical design

1. Introduction
Stats NZ has been using administrative data in the production of economic
statistics for several decades. However, prior to 2015 administrative data
sources have been predominantly used to supplement survey data. In recent
years, a number of our key economic outputs have moved towards an
‘administrative data first’ paradigm. The development and implementation of
an administrative data first approach for quarterly financial statistics is one of
our major achievements in making greater use of administrative data in
economic statistics.
As figure 1 illustrates, before 2001 Statistics NZ’s quarterly financial
collections used a sample survey design. In 2001, a small amount of Goods
and Services Tax (GST) administrative data (less than 10% or 15% of the sales
value depending on different outputs) was added to the design for the first
time as a supplement to the survey data. GST is a comprehensive ‘value added’
tax collection undertaken by New Zealand’s Inland Revenue Department (IRD).
In the last few years, as part of Stats NZ’s business data transformation we
have developed an approach to maximise the use of GST data in the
production of quarterly business financial statistics. This has resulted a new
design for quarterly financial collections, where GST data (direct or modified

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