ERP re-re-re-revisions – explanation and implications

Australia’s official population count, Estimated Resident Population (ERP),  is always revised following a census. This year however additional revisions have been announced following a methodological improvement in census processes, this will have ongoing implications for all users of ERP data …

Glenn wrote about population revisions last year in his blog post ‘Where have all the people gone?‘, and noted some of the ABS’ methodological improvements in ‘Census Australia 2011 – a quality result?’. This post explains the recently announced additional revisions and points out some of the possible impacts for data users.

In December 2012 the ABS announced, through its Australian Demographic Statistics publication, that they will be revising the ERP back to 1991, with the revised data superseding all previously published ERP data and scheduled to be released mid-2013.

It’s important to note that while the revisions are  part of the normal 5-yearly ERP production cycle, the revision of 1991-2011 ERP is one-off, and the historical revisions are likely to affect users unevenly.

Here at .id we are working constantly to prepare population forecasts with the most up-to-date data available, this means we are used to riding the waves of provisional, final and revised population data. Change is constant as they say,  here is a tongue-in-cheek suggestion for the ERP soundtrack, should the ABS ever expand their outputs. That said, the upcoming revisions are significant and are likely to affect a number of our key products.

Revisions are part of the normal 5-yearly cycle…

Estimates of the number of people in a given area are based on the 5-yearly census counts which forms a “base” for the subsequent population estimates. After each census a new base is created and the population estimates from the previous 5 years are updated (“rebased”). As part of the rebasing process intercensal error is identified. Intercensal error is the difference between rebased and the unrebased population estimates (ERP) and it’s calculated by subtracting the newly rebased ERP  from the ERP based on the previous Census which has been updated using the components of population change (births, deaths, and migration).

The revision of 1991-2011 ERP data is a one-off

The revision of the historical ERP, back to 1991, has been identified by the Australian Bureau of Statistics (ABS) as being necessary following a methodological improvement in census processes. The  feature article within the Australian Demographic Statistics publication (cat. no. 3101.0),  ‘The 2006-2011 intercensal period and revisions to historical ERP‘,  gives an explanation of the need for revisions as well as other information relating to the user consultation the ABS has undertaken, the decision made, methods to be used in revision, and the expected impact of the revisions.

In short, the ABS has now decided to revise historical ERP back 20 years to 1991 as the methodological response to the relatively high 2006-11 intercensal error. Historical ERP estimates from September 1991 to March 2011 will be revised through this process.

Revisions will affect users unevenly

As Glenn noted in his blog ‘Where have all the people gone?’, in the 2011 rebasing process some areas have seen very large revisions downwards, to the point of losing most of their five-year population growth. In other areas the ABS revised population estimates upwards in many areas, predominantly in the inner cities, many parts of Perth, and in the mining areas of WA. The impact of these revisions will affect users unevenly because the error varies by area. We’ve already observing significant health cuts in both Queensland and Victoria as a direct result of the downward revisions of each of these state’s ERP.

The ABS have indicated that revisions will have the largest impact on 2006 ERP estimates and the smallest impact on 1996 ERP estimates. What is not yet known is the scale of changes at the state and territory, and sub-state geographic levels. As many data users already know, the usual 5-yearly post-census revisions can result in big changes to the official population count of a small area.

The revisions to historical ERP will have implications for other ABS demographic data, including fertility and mortality rates.  A summary of the series for which revisions will be required will be included in the 20 June 2013 issue of Australian Demographic Statistics (cat. no. 3101.0).

The revisions to historical ERP data will also need to be applied to other ABS data for which historical ERP is a key input. Two of the most prominent examples of these data include Labour Force and National Accounts.

The feature article within the Australian Demographic Statistics publication; a  Basic Guide To Revisions To Historical Population Estimates in 2013  is worth reading as well as it explains the rationale for the revisions and the timing of data releases next year in more depth than this blog.

How have the previous census revisions affected your LGA or community? What have been the flow-on effects of these changes? How do you expect the upcoming revisions to affect your local area’s historical ERP? Feel free to leave a comment below.

Access .id’s demographic resource centre to find ERP figures in the new profile.id sites as well as other population statistics for Australia, States, Capital Cities, Local Government Areas and suburbs.