Page 370 - Contributed Paper Session (CPS) - Volume 2

CPS1888 Chen C. et al.
directly collected), which can provide reliable benchmarks or useful auxiliary
information to the GST data use. The method is mainly applied to small and
medium-sized GST groups where employment data can be used to apportion
GST data to the individual group members. As quality of the apportionment
process cannot be easily measured, we constrain the use of the method to less
than 5 percent of total industry value.
Other sources approach - This method is used when the GST data is
deemed as not suitable for use, therefore non-administrative data sources are
used to produce statistical output, e.g. a quarterly managed collection. This is
further discussed in the next section.

2.3 Statistical Challenges
The major issues that needed to be resolved was for businesses where GST
use was not suitable, and variables are required that are not in GST or other
administrative datasets.
Managed collection strategy - The strategy was developed to provide a
framework that identifies businesses in which administrative data are not
suitable for our statistical use. These businesses are included in a Stats NZ
quarterly managed collection where we will continue to collect key economic
variables that are required on a quarterly basis. A managed collection can be
established using three guiding principles: significance, dominance, and
complexity. For example, the business rules implemented in the current
quarterly outputs are listed below.
• A $100-million significance rule - if an enterprise, or group of enterprises
linked by ownership, have an annual GST turnover of more than $100
million.
• A 3 percent industry dominance rule - if an enterprise makes more than a
3 percent contribution to annual total income for an industry.
• A structure complexity rule - all enterprises that have a significant level of
activity across more than one industry.
Approaches for missing variables - A statistical challenge we had moving
to the ‘administrative data first’ approach is how to produce statistical outputs
for variables not in administrative datasets, such as inventories (or stocks).
Quarterly inventory measures are required for the National Accounts (NA) in
the production of GDP, but there is no quarterly inventories data available in
GST or other administrative sources. We established several methods for
measuring the missing variables. The ones have been implemented to produce
inventories include:
• Benchmark to annual approach – estimates are obtained by ‘rating up’ the
aggregate managed collection inventory series using annual financial
data. This method is applicable where the managed collection capture

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