Inside the sausage machine: The budget is still unfair
What makes Hockey's analogy particularly striking is its applicability to this year's budget process. While the government threw away the very toughest bits of gristle from last year, a number of the most unpalatable cuts are still in the mince, plus some added sweeteners.
Like last year, we have made some calculations showing the impact of the budget measures on disposable income in July 2017, once most of the proposed indexation pauses have taken effect.
Our assumptions are conservative. We consider as status quo the repeal of changes to income tax rates and the low-income tax offset. Like last year, we do not factor in the abolition of the Schoolkids Bonus, or the Income Support Bonus, because this was not strictly speaking a budget measure.
Restricting eligibility for Family Tax Benefit Part B, or FTB-B, may lead to substantial losses of disposable income for families with school-aged children - even before the Schoolkids Bonus is taken away.
Our figures show that a couple with two children aged 11 and 8, where one parent earns A$60,000 per year, would lose A$84.43 per week, or 7.4% of disposable income.
A single parent with one child aged 8 and no private income stands to lose A$49.93 per week, or 10.9% of disposable income.
Pausing indexation of all FTB payment rates affects the most vulnerable families. A couple with no private income and one 3-year-old child would lose A$11.24 per week, or 1.8% of disposable income, while a single parent would lose A$8.80 per week, or 1.6% of disposable income.
Working families on modest wages face a double hit if indexation pauses apply both to payment rates and thresholds. A couple with one child aged 3, where one parent earns A$60,000 per year, would lose A$21.86 per week, or 2.1% of disposable income.
The same family with two children aged 6 and 3 would have A$27.81 per week less to spend, a loss of 2.4 % of disposable income. The losses for a working single parent are A$20.75, or 2%, and A$26.69, or 2.3% respectively.
Families with teenagers will also forgo indexation and receive no compensation for the wind back of FTB-B.
For a single parent with one child aged 14, this means a loss of A$63.70 per week - 13.4% of disposable income if the parent is unemployed and 7.4% on an income of A$40,000. A couple on income of A$60,000 with a 14-year -old could lose up to 79.61 per week, or 7.5%.
These figures represent maximum losses of disposable income. Couples may experience lower losses if both members work. Single parents may also have different outcomes if, for instance, their family tax benefits are subject to the maintenance-income test.
Importantly, we do not include the impact of changes to child care, but if families are not currently using child care and do not use it after the changes, then our figures will be a reasonably accurate guide to the impact on these families (for example, those with school age children not using after-school care).
What NATSEM measured
Our figures broadly agree with the cameo analysis produced by NATSEM, when tax changes and the Schoolkids and Income Support Bonuses are taken into account.
The NATSEM microsimulation model comes to the fore, however, in its ability to model the overall impact of complex policy changes such as the Child Care Subsidy, and its estimation of distributional impacts for the whole population - not just selected family types.
The childcare package is the centrepiece of the budget for households. It is estimated to cost A$4.4 billion over 4 years. In isolation, the package appears progressive and increases assistance more for low and middle income families than for higher income families, with the subsidy for childcare costs reducing from 85% to 50% as family incomes rise.
To finance these reforms the government proposes to maintain some initiatives from the 2014-15 budget.
These include freezing family tax benefit (FTB) rates for two years, adjusting supplements linked to the benefits and freezing the upper income test threshold so that more people lose payments as their incomes increase, and most significantly to stop paying FTB Part B when the youngest child turns six.
There is uncertainty about the overall size of these savings. Because these changes were factored into last year's budget they are not identified as new measures in the 2015-16 budget.
Just before the budget, the Weekend Australian estimated these changes would cut payments by A$9.4 billion over four years. In addition, the government is proposing new changes to family payments and paid parental leave that would save more than A$1.6 billion over four years.
Clearly, the total volume of assistance for families is going down. To assess the overall household impact of the budget, it is necessary to balance who wins from the generally progressive child care assistance proposals versus who loses from last year's and the new savings proposals.
NATSEM analysed the impact of much more than the changes in family assistance and child care and include 25 changes in the first two Abbott government budgets, comparing these with what would have happened if the previous government's policy parameters had been unchanged.
This distributional analysis involves modelling policy changes for some 45,000 real families included in two years of the Australian Bureau of Statistics Survey of Income and Housing.
NATSEM produces distributional impacts for quintiles (20%) of households by family type - couples with and without children, lone parents and single person households.
Both couples with children and lone parents lose on average, with the poorest quintile of couples losing just over A$3,000 a year or 7.1% of their disposable income and the poorest quintile of lone parents losing just under A$3,000 a year or around 8% of their disposable income.
Most households without children - except the poorest 20% - are estimated to have minor increases in real disposable incomes by 2018-19.
The government in Question Time has emphasised that the NATSEM calculations do not include any "second round" impacts of the budget changes.
That is, the policy package put forward by the government makes work more attractive both by reducing the cost of childcare but also by cutting benefits to families, giving them greater "incentive" to increase their hours of work to make up for the loss of FTB-B in particular.
Will the Budget increase workforce participation?
Asked about the modelling during question time, the prime minister said this omission meant the modelling was "a fraudulent misrepresentation" of the government's budget because returning people to work was "the whole point of the policy measures".
At one level, this sounds like a reasonable criticism. The explicit aim of the budget changes is to make increased hours of work more attractive to families.
However, Treasurer Joe Hockey has also conceded that "as a rule second-round effects are not taken into account" in any budget.
This is because while there are likely to be some behavioural responses to these changes, the size of that response is unclear.
A 2007 Treasury Working Paper points out that estimates of labour supply responses to tax and benefit changes can vary widely.
The Productivity Commission in its report on childcare that formed the basis of the proposed childcare changes in the budget was cautious about the size of the labour supply response to its recommendations, arguing that additional workforce participation will occur, but it will be small, and is estimated to increase the number of mothers working by 1.2% (an additional 16,400 mothers).
It is also worth pointing out that the economic parameters underlying the overall budget suggest that employment effects are not likely to be substantial.
The labour force participation rate is projected to rise marginally from 64.6% to around 64.75% over the forward estimates, but the unemployment rate is projected to increase from 5.9% to between 6.25% and 6.5%, which implies a small fall in the proportion of the adult population who are actually employed.
Overall, while there will be some second-round positive effects it is highly unlikely that they will offset the losses in disposable income experienced by many families with children.
Governments should welcome the type of evidence-based policy analysis exemplified by NATSEM's work, and ideally provide it themselves. It focuses the debate on concrete questions of how policy changes affect people's lives.
To criticise the straightforward modelling approach because it yields the "wrong" answer smacks of shooting the messenger. The government should be upfront with the public about exactly what is in the budget sausage.
Peter Whiteford is a Professor with the Crawford School of Public Policy at Australian National University
Daniel Nethery is a PhD candidate with the Crawford School of Public Policy at Australian National University