Column

Internal migration significantly impacts local population changes, influencing growth or decline, as well as population age-sex compositions. Understanding shifts in internal migration levels, patterns, and compositions in response to economic and housing market conditions, natural disasters and pandemics requires up-to-date and reliable time series data on the movements. Accurate data that are recent in time poses a significant challenge, particularly in countries lacking population registers that compel people to report all changes in address. In countries relying on censuses or other administrative sources, migration statistics may be too delayed to offer insights into evolving migration patterns. In this context, novel datasets incorporating digital traces emerge as possible complements to traditional data sources. Their advantages lie in their spatial granularity and continuous coverage, which could add insights into the dynamics of internal migration, especially during crises.

The COVID-19 pandemic significantly disruptive the pattern of internal migration flows in Australia, leading to net gains in regions and losses from greater capital city regions. Our understanding of the geography and timeline of these disruptions is, however, limited due to a lack of disaggregated internal migration data in non-census years. In this paper, we generate annual estimates of in, out and net flows for SA4s Australia for period 2019 to 2022 using data from Muval, an online removalist website. We estimate the relationship between Muval data and internal migration estimates obtained from the 2021 Australian Census of Population and Housing in a multiple regression framework and use this relationship to predict SA4 migration for other years. The results confirm significant volatility in the distribution of SA4 flows, peaking in 2020 at the height of Australia’s COVID response. While the Capital City SA4s remain the dominant source and destination of migrants throughout this period, there are significant shifts in their order, with out flows from SA4s in Melbourne and Perth decreasing in relative terms, and in flows to Darwin increasing. As expected, the relative share of inflows also increased to a number of attractive peri-urban SA4s. The results provide the first spatially disaggareted pricture of intervnal migration pattersn over the COVID period, with the proposed framework also allowing the continued “nowcasting” of internal migration in advance of official ABS releases.

About This dashboard has been developed and is maintained by a team of researchers of the Queensland Centre for Population Research at the University of Queensland.