Economics says Australia should have negative migration

Politics has forced the hand of the major parties to announce immigration cuts.

But on those promises, Australia would run one of the highest per-capita intakes in the developed world – above most of Western Europe, where immigration is fracturing the social fabric. Politicians imagine themselves trapped between an economy needing immigrants and a public wanting fewer. It's time they questioned their assumptions.

Lobbying by education providers for international students should also be dismissed. Louie Douvis

Start with productivity. Our lacklustre productivity performance has coincided with a migration program that is less skilled than advertised. Many immigrants work below their qualification level, while the permanent "skilled" program is less skilled than its name suggests: only 46 per cent were primary applicants – the rest were spouses and children.

Across the skilled, family and humanitarian streams, Treasury puts the lifetime fiscal dividend at just $41,000 a head. Even that rests on an infrastructure cost of only $40,000 to $70,000 per immigrant. University of Queensland's Jane O'Sullivan puts the infrastructure requirement at no less than $100,000 per immigrant. On that estimate, much or all of Treasury's dividend disappears.

The main evidence that Australia's immigration program makes us more productive is weak: a series of Treasury-funded OECD papers, one of which it headlined as "Migrants boost the labour productivity of Australian-born workers".

The paper is more circumspect. Immigrants move to places where wages and productivity are already high, making it difficult to tell whether immigration causes prosperity or follows it. The authors try to separate the two, but the result depends heavily on statistical assumptions, and some findings ran opposite to expectations. Taken at face value, the paper's conclusion states that immigrants are "associated with" higher-wage regions – what you'd expect when they cluster in Sydney and Melbourne.

More fundamentally, the economic case for continually expanding labour supply is increasingly challenged by frontier research. Daron Acemoglu – the 2024 Nobel laureate – finds that ageing countries have grown faster, plausibly because scarce labour encourages firms to adopt labour-saving technology. This is intuitive to anyone who has watched automated car washes give way to hand-wash bays staffed by immigrants – yet lost on our nation's economists, now accidental Luddites.

Even if ageing-induced labour scarcity were a problem, immigration is a lazy solution. Australians are living longer, yes – but also healthier. If over-65s worked at New Zealand's rate – 26 per cent against our 16 – we would gain half a million workers. Experienced, English-speaking and, unlike new immigrants, already housed. Removing 80 per cent effective marginal tax rates on pensioner earnings is a no-brainer.

Immigration also enables reform-shy governments. Australia has a doctor shortage but no shortage of school-leavers desperate to become them. Yet we ration medical school and specialist training places, then import overseas-trained doctors. Meanwhile, artificial intelligence models outperform doctors on diagnostic tasks, creating opportunities for nurses to take on work reserved for doctors. Healthcare needs a shake-up that industry leaders will never volunteer – it must come from bold political leadership.

Other labour shortage arguments are weaker still. We are told immigrants are needed to build houses – for demand that immigration creates – but the program admits almost no construction workers. In aged care, the clients are the wealthiest cohort in Australian history and Treasury's Retirement Income Review concluded that retirees "die with the bulk of the wealth they had at retirement intact". Australian-born workers were three-quarters of the aged-care workforce in the 1990s – they are now a minority. Australians would wipe 90-year-olds' bums if wages weren't set to insulate bequests.

Lobbying by education providers should also be dismissed. The Australian Bureau of Statistics estimates $54 billion of "education exports" but only $23.5 billion is tuition. That $30 billion difference is modelled consumption – an implausibly high figure implying students from predominantly developing countries are spending roughly $50,000 annually on living costs. Meanwhile, the ABS counts $15.6 billion of wages to students – two-thirds of the fee take. It cannot see cash-in-hand work, or money moving internationally through hawala, suitcases and cryptocurrency. Correct for what's missing, and education is likely a much smaller export.