MEASURING THE STATISTICAL UNDERGROUND


SCOPE AND RATIONALE

7.1 Although the 2008 SNA does not discuss the statistical underground, the OECD Handbook suggests that production missed due to deficiencies in data collection (for example under coverage of enterprises, non–response, and under reporting) is as much a part of the NOE as the other NOE components.

7.2 There are no explicit adjustments made when compiling the ASNA to cover the statistical underground. ABS data sources are subject to a quality assurance process that:

• closely monitors and adjusts the data sampling and collection frames to ensure they are up–to–date,
• employs a rigorous follow–up of survey and data collection non–responses to ensure a high level of data coverage, and
• subjects all data responses to validation checks to ensure the correctness and consistency of data.

7.3 To the extent that source data are deficient, compilers of ASNA seek supplementary and/or alternative sources. For example, some higher quality data sources are not available with the frequency needed for account compilation. In such cases less reliable short–run indicators are used to move estimates based on the more reliable, but less frequent sources.

7.4 Because of the variety of techniques used to manage source data quality, there are no overall measures of the statistical underground. However, it is worth commenting on two recent quality improvement exercises undertaken by the ABS.


UNDOCUMENTED IMPORTS

7.5 The ABS collects and publishes monthly estimates of the value of Australian international trade in goods and services. These statistics provide information about the value of Australia's trade with the rest of the world and should include the value of all goods and services that are purchased by residents from non–residents. These statistics are used for a range of purposes, including the estimation of GDP and Balance of Payments. It is important that any deficiencies in the coverage of these statistics are addressed to ensure their ongoing relevance. There are two related coverage issues:

• goods entering the country and not being documented at the customs border, and
• intangible goods and services supplied electronically and not subject to border clearance.

Undocumented goods imports

7.6 The ABS compiles international merchandise statistics using data submitted by exporters and importers, or their agents, to the Australian Customs and Border Protection Service (Customs and Border Protection). The coverage of this data is limited to goods that cross the customs frontier and for which detailed customs clearances are required. To facilitate the efficient flow of goods and enhance the cost effectiveness of Australian tax and duty collections, goods valued below the low value threshold (LVT) are exempt from tax collection and detailed import declarations. In November 2005, this threshold was increased from $250 to $1,000 to further reduce compliance costs.

7.7 The relatively high LVT creates an inherent coverage problem. The change to a higher threshold, as well as changes in consumer behaviour (such as direct purchase over the internet), suggests that there may have been significant recent growth in the value of goods imported into Australia below the LVT.

7.8 In August 2013 the ABS published the Information paper Measurement of Online Retail Trade in Macroeconomic Statistics, 2013 (cat. no. 8501.0.55.007) that contained an improved estimate for imports of goods under the LVT. The ABS has been working with Customs and Border Protection and Australia Post (Auspost) data to attempt improved estimates for imports of goods under the LVT. The modelled estimate of $6,226 million for 2011-12 was based on counts of items arriving in Australia as recorded by Auspost, some data about average parcel values provided by Customs and Border Protection, and data about items being imported from market research sources. It is expected that a complete set of estimates for the LVT will be included in the Balance of Payments coverage adjustment for the first time in the August, 2013 issue of International Merchandise Imports, Australia (cat. no. 5439.0).

Electronic intangible goods and services imports

7.9 Electronically imported intangible goods and services such as music, movies and software downloads, international transport, and accommodation are increasing in volume. Traditionally, such imports were largely undertaken by businesses and are measured by the ABS quarterly Survey of International Trade Services (SITS) – a business based survey.

7.10 However, households are now directly importing such services in greater numbers, and it is impractical to include households in the SITS. While it is possible to ask relevant questions in surveys of households, the most likely survey, the Household Expenditure Survey, is costly and is conducted only every six years. This frequency of data collection is insufficient to measure trends in this developing market.

7.11 A number of private sector research efforts are attempting to measure eCommerce in its many manifestations, and some of these have been able to produce estimates of household imports using methods that analyse data from payments systems. The ABS is examining the feasibility of using this research to estimate missing data for electronic deliveries and validate modelled estimates for LVT goods imports. The results of this work are expected during 2013.


BUSINESS REGISTER COVERAGE AND MAINTENANCE PRACTICES

7.12 A second quality assurance project was designed to overcome questions about the Australian Business Register (ABR) that is maintained by the Australian Tax Office (ATO) and used by the ABS as the basis for business survey sampling frames. In common with the use for statistical purposes of any dataset designed to support administrative purposes, there is a tension between the different administration and statistical objectives. For example, on a number of occasions the ATO has changed the rules under which an Australian Business Number (ABN) is issued. As the ABS uses the ABN as an indicator of the existence of a business, the application of these administrative rule changes has had an impact on the classifications used in the ABS business statistics.

7.13 The quality assurance project reviewed all aspects of the use of the ABR for statistical purposes, including aligning those uses with the SNA production boundary. A number of improvements were identified and incorporated in operational processes so that the ABS survey frames now reflect the production boundary more accurately.