About This Tool

This app helps Australians understand projected life expectancy based on socio-economic and demographic factors.

It uses data from the Person Level Integrated Data Asset (PLIDA), accessed via the Australian Bureau of Statistics' DataLab. The life expectancies are computed based on the 2016-2017 census population with improvement factors of ALT2015-17 (125 year).

Related Publications

• Fei Huang, Francis Hui and Andrés Villegas (2025). Towards Fairer Retirement Outcomes: Socio-Economic Mortality Differentials in Australia. SSRN Working Paper.
• Fei Huang, Francis Hui and Andrés Villegas (2025). Australian Socio-economic Retiree Mortality 2016-17. Harvard Dataverse. https://doi.org/10.7910/DVN/C0HRDE.
• Pramo Samarasinghe, Fei Huang, Francis Hui and Andrés Villegas (2026). Towards Fairer Retirement Outcomes: Health-Related Mortality Modelling. Actuaries Institute.

Authors of the Publications

Fei Huang, Francis Hui, Pramo Samarasinghe, and Andrés Villegas

Fei and Andrés are Associate Professors at the School of Risk and Actuarial Studies, UNSW Sydney. Francis is an Associate Professor at the Research School of Finance, Actuarial Studies, and Statistics, Australian National University. Pramo was an Honours student at the Australian National University and is now a research assistant at UNSW Sydney.

Acknowledgement

We thank the Australian Government Actuary Guy Thorburn and his team for their invaluable support in this project.

Limitations and Assumptions

The results are estimates from a model. These points describe what the calculation does and does not allow for.

• Independence of the two lives. The couple calculator treats the two lives as independent given the selected characteristics. Other factors that the characteristics do not capture, such as shared health and lifestyle, may link the two lifespans. If the lifespans are positively linked, the time until the second death may be shorter than shown.
• Characteristics held fixed. The selected characteristics are held fixed throughout the calculation, for one person or for a couple. If one partner dies, the survivor's mortality is not recalculated using the rates for people who are not currently married, a group that includes widowed people. In the model, mortality rates for people who are not currently married are generally higher than those for married people, so the time until the second death may be longer than shown.
• The 2016 base year. The cohort basis is anchored to the 2016 base year of the data. A person of the selected age is treated as if they were that age in 2016, and assumed future mortality improvements are applied to the years that follow. Any change in mortality between 2016 and today is not included.
• Ages beyond 100. The model is fitted to data up to age 100. Life expectancy, and the average time until the second death, include years lived beyond 100, which rely on extrapolated mortality. The 1 in 5 and half figures that are below 100 do not depend on them. Ages of 100 or more are shown as 100 or older, because the data end there.
• Postcodes. A postcode result uses the IRSAD decile of the whole postal area, which is a general guide. The mortality model uses deciles of smaller local areas.
• Marital status. The data use registered marital status. Never married, widowed, divorced and separated people form one group, and people in a de facto relationship are in that group too.

Disclaimer

This tool provides general statistical information about life expectancy for groups of people with the characteristics you select. It is not a prediction for any individual and does not take account of personal circumstances such as health or lifestyle. It is not financial, retirement, insurance, medical or legal advice and should not be the only basis for any decision. Estimates are based on 2016–2017 data and modelling assumptions, and carry statistical uncertainty. Before making decisions that depend on this information, consider speaking to a suitably qualified professional.

Project Lead Contact

Fei Huang (feihuang@unsw.edu.au | www.feihuang.org)

Download Data

You can download the modelled period-based probability of death (qx) for ages 60–100, covering all socioeconomic profiles below from the Harvard Dataverse, in the Australian Socio-economic Mortality Dataverse.

⬇ Download Dataset

How Did This Help You?

We’re interested in understanding how this tool supports real-world decisions, insights, or research. Whether you're a retiree, policymaker, educator, student, advisor, or researcher, your experience helps us improve this tool and demonstrate its impact.

If this tool helped you explore a scenario, inform a conversation, support a policy submission, or improve your understanding of retirement outcomes, we’d love to hear your story.

You can share as much or as little detail as you like. The form is optional and anonymous unless you choose to include your name or email. With your consent, we may include anonymised stories in research reporting or impact assessments.

💬 Open the feedback form

The form opens in a new tab. You can also contact us directly at feihuang@unsw.edu.au.