bs_bs_banner Asia & the Pacific Policy Studies, vol. 3, no. 2, pp. 249–265 doi: 10.1002/app5.135 Original Article Implications of Australia’s Population Policy for Future Greenhouse Gas Emissions Targets Corey J. A. Bradshaw* and Barry W. Brook Abstract 1. Introduction Australia’s high per capita emissions rates Australia is the world’s sixth-largest country makes it is a major emitter of anthropogenic (land area = 7.69 million km2), yet it has a greenhouse gases, but its low intrinsic growth 2014 human population of only 23.5 million, rate means that future increases in population making it the 51st largest national population size will be dictated by net overseas immigra- in the world (worldbank.org), or approxi- tion. We constructed matrix models and mately 0.3 per cent of the planet’stotalhuman projected the population to 2100 under six dif- population. Despite this relatively small popu- ferent immigration scenarios. A constant 1 per lation, Australia has one of the highest cent proportional immigration scenario would per capita greenhouse gas emissions rates in result in 53 million people by 2100, producing the Organisation for Economic Co-operation 30.7 Gt CO2-e over that interval. Zero net im- and Development, because of its heavy migration would achieve approximate popula- reliance on coal-fired and gas-fired electricity tion stability by mid-century and produce 24.1 generation, an expansive fossil-fuelled transport Gt CO2-e. Achieving a 27 per cent reduction in network, and large agricultural sector annual emissions by 2030 would require a 1.5- (International Energy Agency 2014). Australia to 2.0-fold reduction in per-capita emissions; is also a major producer of fossil fuels, having an 80 per cent reduction by 2050 would re- exported approximately 11,600 petajoules quire a 5.8- to 10.2-fold reduction. Australia’s (PJ) of primary energy in 2013, of which ~80 capacity to limit its future emissions will there- per cent was coal and ~10 per cent was natural fore depend primarily on a massive technolog- gas (abs.gov.au). When combusted, this ical transformation of its energy sector, but equates to approximately 1.3 per cent of the business-as-usual immigration rates will make world’s total anthropogenic greenhouse gas achieving meaningful mid-century targets emissions (Brook 2012a). more difficult. In 2007, Australia committed to reducing its greenhouse gas emissions by ratifying the Kyoto Protocol (United Nations 1998) and Key words: demography, fertility, dependency signing the second commitment period ratio, emissions, climate change (2013–2020) (Bradshaw et al. 2013). Australia’s current pledge is to reduce its emissions by 5 per cent of its year 2000 National Greenhouse Gas Inventory (NGGI) * Bradshaw: School of Biological Sciences, University of Adelaide, Adelaide, South Australia total by 2020 (dfat.gov.au). In the Clean 5005, Australia; Brook: School of Biological Energy Act 2011, the government of the day Sciences, University of Tasmania, Hobart, Tasmania had set a reduction target of 80 per cent of 7001, Australia. Corresponding author: Bradshaw, 2000 emissions by 2050, but that was repealed email <[email protected]> (Commonwealth of Australia 2011). Since © 2016 The Authors. Asia and the Pacific Policy Studies published by Crawford School of Public Policy at The Australian National University andWiley Publishing Asia Pty Ltd. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. 250 Asia & the Pacific Policy Studies May 2016 then, a 26 to 28 per cent reduction below 2005 2005 emissions) and a putative 2050 target of emissions (equivalent to 21 to 23 per cent be- an 80 per cent reduction (from 2000). We low 2000 emissions) by 2030 has been set ask broadly how effective population policies (Australian Government 2015). After 2030, can be in mitigating Australia’sfuture Australia’sofficial commitment is still greenhouse gas emissions (both cumulative undecided. How Australia will manage to outcomes and annual rates), and explore achieve the 2030 target, and any longer-term the sensitivity of this conclusion to net goals such as an 80 per cent reduction by immigration policy. We also quantify the 2050 has been the subject of many studies, major economic outcomes—in terms of child, mainly focused on increasing efficiency aged and health care costs—of the changing and penetration of renewable electricity population age structures under various generation (Beyond Zero Emissions 2010; immigration scenarios. Seligman 2010; Elliston et al. 2012; Palmer 2012; Trainer 2012; Australian Energy 2. Methods Market Operator 2013; Denis et al. 2014); however, few have considered the direct 2.1. Demographic Data impact of an increasing Australian population on meeting these targets (although see We obtained life table data (age-specific Brook 2012a). mortalities and fertility from 0 to 100+ years As a wealthy nation (that is, world’s of age) for the Australian population from the sixth-highest per capita gross domestic Australian Bureau of Statistics <http//:www. product = US$67,463; worldbank.org), abs.gov.au>. We converted the aggregated Australia’s demography is typical of economi- 5-year age class births per 1000 women into cally developed nations in that its intrinsic fer- age-specific fertilities (mx) by dividing the tility is below replacement (average number of 5-year classes equally among their constituent children born to a woman who survives to the years and accounting for breeding female end of her reproductive life = 1.78; replace- mortality within each of the 5-year classes ment = 2.1) (CIA World Factbook 2011). As (Bradshaw & Brook 2014). The Australian such, its current population growth is dictated Bureau of Statistics also provides age-specific mainly by net overseas migration (Turton & (x) yearly population estimates (nx)from Hamilton 1999). Accumulated greenhouse 1971 to 2014. These estimates are obtained gas emissions are strongly related to popula- by adding to the estimated population at the tion size (Shi 2003), so it stands to reason that beginning of each census period the compo- any policies to reduce national emissions nents of natural increase and net overseas should also incorporate population projections migration (www.abs.gov.au). All age-specific in their assessments. However, a critical over- population size, mortality and fertility data we view of the contribution of population growth derived are available online at DOI:10.4227/ to achieving both its short-term and longer- 05/55679E714245D. term emissions-reduction targets is lacking or obsolete (Foran & Poldy 2002). 2.2. Leslie Matrix We address this gap by producing a compre- hensive demographic model of the Australian We defined a pre-breeding, 100 (i)×100 (j) population with projections to 2100, assuming element, Leslie projection matrix (M)for various rates of future net overseas immigra- women only, multiplying the subsequent tion. Based on these projections, we forecast population vector by the 2014 stage-specific business-as-usual and zero-immigration sex ratio to estimate total population size at emissions trajectories to calculate the per capita each forecast time step (Bradshaw & Brook reductions required to meet the 2020 2014). Fertilities (mx) occupied the first row emissions-reduction target (5 per cent), the of the matrix (ages 15–49 years), survival median 2030 target (27 per cent reduction of probabilities (1 À Mx) were applied to the sub- © 2016 The Authors. Asia and the Pacific Policy Studies published by Crawford School of Public Policy at The Australian National University andWiley Publishing Asia Pty Ltd. Bradshaw & Brook: Australia’s Population Policy Implications 251 diagonal, and the final diagonal transition increasing number of environmental refugees, probability (Mi,j) represented survival of the for example. Scenario 4 simulated a zero- 100+ stage. Complete R code (R Core Team immigration policy (no net overseas migra- 2014) for the scenario projections is available tion), whereas Scenarios 5 and 6 simulated from the authors upon request. fixed annual net immigration at 20,000 and 100,000, respectively (Table 1). 2.3. Immigration It is arguably unrealistic to assume that the demographic rates (survival, fertility) would We obtained net overseas migration data from remain stable from 2014 to 2100 especially 2004 to 2013 for women and men and their noting recent trends. We therefore repeated 5-year age class structure from the Australian all scenarios assuming a continuous (linear) Bureau of Statistics. We applied the average increase in average age of (female) breeding age structure to a migration vector constructed (increasing average age of primiparity) by allo- for each of the migration scenarios (Figure S1 cating 50 per cent of the fertility in the youn- and see following discussion) and added this gest reproductive age class (15–24 years) to the population vector for each yearly evenly across the older breeding classes iteration of the projections. (25–49 years), following a linear change func- tion from 2014 to 2100 (Bradshaw & Brook 2.4. Projection Scenarios 2014) (Table 1). According to the Australian Treasury’s 2015 Intergenerational Report For each projection, we multiplied the Nx (Commonwealth of Australia 2015a), life vector by M for 86 yearly time steps (2014 to expectancy is predicted to increase from 93.6 2100). All projections were deterministic. to 96.6 years for woman from 2015 to 2055. We applied a broad range of immigration This represents a 3.2 per cent increase in scenarios to examine the effects of various average survival, so we also conservatively immigration policies on long-term population estimated a 3.2 per cent reduction in mortality change and their associated emissions profiles across all age classes achieved linearly by (compared with Turton & Hamilton 1999; 2100 (Table 1).