Moderate changes to Australia's social support programmes could deliver better outcomes for those receiving assistance and result in billions of dollars in budget savings, according to a paper published by actuary Hugh Miller.
For each vulnerable individual, social service spending across welfare, housing, health care, justice and child protection can easily amount to A$500,000 ($361,000) to A$1m over a lifetime, Dr Miller said in his thought leadership paper “People, projections and payments: A look at modern government service delivery” which has been published by the Actuaries Institute.
Australia’s local, state and federal governments have clear opportunities to speed up improvements, he said.
“Most importantly, a large portion of the savings comes from better outcomes for individuals who might move into active employment, sustain their housing situation, or reduce their criminal offending. This means they have better options for a fulfilled life,” he said.
Dr Miller argues that while there is much good work already going on, in many ways “governments can do better”.
“The large amount of money being spent, plus the significant potential impact on people’s lives means that even incremental improvements in government services will yield substantial benefits over time,” the paper states.
Dr Miller said Australian governments will face growing budget pressures, including the demands made by an ageing population. Other pressure points include rapid growth of disability pension payments, increases in public housing (in New South Wales in 1960 only 15% of public housing tenants were on government benefits compared with 94% today), rising health care costs and a continuing increase in prison populations despite falling crime rates.
“Governments can improve their ability to target recipients, intervene earlier and use innovative funding mechanisms to support ongoing programmes,” he said. “These goals will only be met with stronger and better links between data sets already collected by State, Territory and Commonwealth authorities, attention on collecting data on people’s broader outcomes, and transparency when it comes to evaluating programmes.”
In Australia, data collection is piecemeal and there are significant barriers to linking together important information across health, housing, education and disability support. The State and Commonwealth divide can also hinder good analysis. “We could use data to see long-term patterns of service use, including people who require extended support over long periods of time,” Dr Miller said, citing early intervention and better targeting of services. This provides significant impetus to focus on prevention to improve people’s subsequent pathways.
“Not every policy experiment will be successful,” Dr Miller said “which means that publishing an evaluation can open policymakers up to criticism and potentially reduce the likelihood of trying new things. Political risk aversion is natural but means that good analysis is often not published. We should not be afraid of the truth.”
He argues that some areas of government already are using data to triage people according to need, such as predicting which job seekers are likely to secure work or who will end up homeless. “This means we can anticipate outcomes good or bad.”
While using predictions to offer different support to people raises important ethical issues, many people are comfortable with the notion of providing higher levels of support to those at greatest risk of poor outcomes.
He adds that equality of provision of services does not guarantee good outcomes for all.
“Children from low socio-economic backgrounds require significantly more support to achieve comparable educational outcomes to children from wealthy backgrounds,” Dr Miller cited as an example.
With improved data linkage, better outcomes data collection, tackling of fragmented services and further improving transparency when programmes are evaluated – all in a way that respects people’s privacy and builds trust with the community – government service delivery can be significantly improved.