REALISING THE POTENTIAL OF NEW DATA SOURCES AND METHODS

As a part of our drive to meet demands for increased and improved statistical services, we are moving beyond traditional data collection approaches to take advantage of new and exciting data sources and methods. These new sources and methods will offer users of statistics the following benefits:

• less burden for survey respondents - an increased use of alternative data sources will reduce the need to collect data directly from respondents
• less costly - reducing the costs of existing approaches will enable the ABS to produce more statistics with existing funding
• more comprehensive - using new methods and a wider range of data sources will enable users to more easily pinpoint and analyse emerging trends
• richer statistics – which are more timely, relevant, accurate, coherent, interpretable and accessible, for meeting policy and research requirements.

By tapping into the potential of these new approaches, and making it part of the core, mainstream production of statistics, we will be able to leverage the power of multiple data sources and apply more sophisticated modelling methods. These new sources and methods will enable us to meet those needs that are beyond traditional data collection approaches.

Linked Employer-Employee Database (LEED)

The ABS is already using new sources and methods to better use data that has already been produced as part of government agencies’ day to day transactional and administrative activities. For example, the foundational Linked Employer-Employee Database (LEED) project, which predominately used business and personal tax data administered by the Australian Taxation Office, to produce new experimental data to inform the Australian labour market.

The LEED foundation projects demonstrate the ABS' commitment to innovate and deliver new information on the Australian labour market using administrative data. For example, through integrating employer and employee data, we are able to measure for the first time that 1.9 million employees were multiple job holders (individuals with two or more concurrent jobs) in 2011-12. These statistics can only be produced by linking employer and employee data.

The foundation projects represent an important step towards the development of a future LEED. The development will address a longstanding information gap in Australian labour statistics by providing a single database capable of addressing complex questions about employer and employee relationships, at both a point in time and across a number of years.

The future LEED will provide a better range of information on labour market dynamics, firm level productivity and a range of regional analytics. New information will also be available to address other complex policy and research questions with no increased regulatory or survey burden.

The long term vision is to extend the LEED to include data sources other than just tax data. Due to its size and high population coverage, a tax-based database has the potential to be enlarged with other administrative datasets and ABS collections (i.e. person and business level survey data, and Census). This will help broaden the information base on the Australian labour market and help address complex policy problems. It also complements the current labour statistics produced by the ABS.

“LEED will enable us to explore, in a very tangible and useful way, a number of key dimensions of the statistical transformation:
• increasing the solution-orientation of the ABS
• realising the benefits of new methods and approaches
• increasing the use of administrative data.”

Michael Smedes, Program Manager, Transforming Economic Statistics, Australian Bureau of Statistics.


Enhancing Australia’s CPI

The ABS has used new methods to enhance one of Australia’s best known statistics, the Consumer Price Index (CPI). The CPI measures changes in the price of a fixed basket of goods and services acquired by consumers in metropolitan private households. This is intended to measure the overall inflation in prices of goods and services acquired by Australian households.

Price data collection was traditionally the role of ABS officers in all capital cities, but the availability of transactions data direct from retailers has now established new opportunities both in the collection and future enhancement of the Australian CPI.

One key aspect of this project was the secondment of Professor Jan de Haan, an internationally recognised prices expert from Statistics Netherlands/Delft University. Professor de Haan was engaged to enable a risk managed approach to CPI enhancement.

The ABS also worked with key stakeholders from the Reserve Bank and the Treasury during the project. Given the interest in enhancing the CPI, this approach was a valuable way for the ABS to supplement its existing expertise and engage with risk in a measured way.

With Professor de Haan’s assistance, the project developed a research program into future enhancements to the Australian CPI. A number of areas were identified where the ABS could further expand its use of transactions data, as well as providing specific advice on methods and business processes.

Project outcomes have led to more efficient use of price and quantity information from transactions datasets and may also be able to enhance other macroeconomic statistics, such as Retail Trade and the National Accounts.

The success of this project so far has encouraged the ABS to increase the use of transactions data. While all new uses will present unique challenges, the methods and processes developed for the CPI will help streamline any implementation process undertaken for other statistics.

There is more work to be done in undertaking one of the biggest changes in the ABS’ history and with the ongoing support of our stakeholders, we are confident we can achieve our vision of unleashing the power of statistics for a better Australia.