This July, we had five public servant scholars graduating from ANU, supported by Sir Roland Wilson scholarships: four PhDs, and one Pat Turner scholar with a Master of Forestry. We are thrilled to see these scholars complete their studies and return to the APS with enhanced knowledge, skills and extensive academic networks to bring back to the service and contribute to evidence-based public policymaking.

Dr Joseph Chien graduated mid-year with a PhD which uses APS administrative data to understand the drivers of productivity. Productivity growth is important for Australian industry so that firms can either produce more at the same cost, or produce the same amount at a lower cost. This gives firms the ability to increase wages, decrease prices (without reducing profits or wages) and increase profits, delivering higher living standards in the long run for all Australians.

Prior to starting his PhD, Joseph was the Director of the Methodology Transformation Branch in the ABS, responsible for data integration, access and confidentiality methodology. Joseph also spent three years in the Economics Directorate at the OECD working on global macro simulation models.

Over his career at the ABS, Joseph has seen an increasing emphasis on using the vast quantity of APS administrative data to produce statistics and insights which inform public policy. While this method of producing statistics offers significant advantages, such as large sample sizes, cost effectiveness and timeliness, it is not without challenges.

Joseph’s research is one example of the myriad ways that administrative data could be interrogated and applied, to answer all sorts of complex policy questions about our society.

One particular challenge is handling large sample sizes when estimating models. In his PhD research, Joseph explored methods to estimate over 12 million coefficients from 130 million observations. These methods enabled him to understand how firm dynamics (ie. firms entering, continuing and exiting the market) contribute to overall productivity.

Joseph showed that firms entering and exiting the market play a key role in improving productivity. This may suggest a need for policies which aim to encourage competition, rather than provide advantages to continuing firms.

Another challenge with using large administrative datasets is applying methods to extract relevant information to understand the complex relationships between firms, and then see how these relate to productivity. In his research, Joseph used a ‘semantic web’ approach to extract complex business network information from integrated administrative data. He combined this information with firm productivity information in statistical network models to describe the factors contributing to firm participation in Australian business networks.

His research showed that larger firms are more likely to form business networks than small and medium-sized firms. This may suggest a need for policies that encourage stronger collaboration between small and medium-sized firms.

The need to maintain confidentiality while making administrative data more accessible and useful for informing policy decisions is always at the forefront of Joseph’s mind. In his research, Joseph explored the use of synthetic data to make business microdata more accessible for research while maintaining confidentiality. Synthetic data is information that is artificially manufactured rather than generated by real-world events. Synthetic data is created algorithmically to approximate the original data, and is used as a stand-in.

Confidentiality of business microdata is a particular challenge in the Australian context as some industries are characterised by an oligopoly or duopoly making some firms easily identifiable. Typical data protection techniques such as information reduction may not be as effective for business microdata.

Joseph’s research showed that synthetic data can work as an alternative approach to maintaining confidentiality while making business microdata more accessible for research. Now that Joseph has returned to the ABS, he is interested in applying the methods he used in his PhD to help the APS make better use of the vast amounts of administrative data that the public service generates.

“This scholarship provides a great opportunity for the public servant scholar, but there’s also a huge benefit for the government sector - to have a scholar bring their close connections with the research community back to the department.”

Joseph is the first person from the ABS to be awarded a SRW scholarship. He has a Bachelor of Commerce from University of Auckland, Grad Cert in Stats from Victoria University, a Grad Cert in Financial Management from ANU, Grad Dip in Economic Policy from the London School of Economics and Political Science, and a Master of Commerce in International Business from UNSW.

Read more about the Sir Roland Wilson Foundation. Keep updated by following us on Facebook and Twitter.

Publication

Chien, C. H., Welsh, A. H., & Moore, J. D. (2020). Synthetic Business Microdata: an Australian example. Journal of Privacy and Confidentiality, 10(2).