Social Income costings

This week I’ve proposed a Social Income, essentially a more generous though means-tested unemployment benefit set at the poverty line of 50% of average income (currently $19,000 a year approximately) and available for people doing a range of socially useful activities, including caring for sick relatives, raising children and volunteering for a registered charity.

For any such proposal, questions of cost immediately arise. Without access to detailed modelling, or payment to undertake such an exercise in detail, it is impossible to be precise as to the likely cost. Below, however, are some extremely rough estimates, as a general guide.

Payments to existing beneficiaries

There are currently 134,000 people on the unemployment benefit (Jobseeker Support), which is worth around $11,000 a year. If they were all willing to undertake socially useful activities while not in paid work, they would each be eligible for $8000 a year extra, at a cost of $1.07 billion. (A few would not be willing, but not enough to make a material difference to this level of calculation.)

There are currently 60,000 people on the former DPB, now known as Sole Parent Support, which is worth around $17,000 a year. Lifting them up to the Social Income would cost $120 million a year.

There are 93,000 people on the Supported Living Payment for those with disabilities and related conditions. It is currently worth $14,000 a year, and lifting them up to $19,000 a year would cost $465 million.

Both the latter two groups currently receive more than the unemployment benefit, in recognition of their higher costs. To maintain the gap between what they receive and the unemployment benefit ($2000 a year in the first case, $5000 a year for the second) would cost an extra $360 million and $279 million respectively.

The total cost of the above would be roughly $2.3 billion. There would, however, be some savings – maybe $300 million, for argument’s sake – in the Accommodation Supplement, on which the country currently spends $1.5 billion.


It is difficult to accurately estimate the cost of extending the benefit to people undertaking care for ill, infirm or disabled members of their family. Reports put the total of informal carers at 430,000.

However, calculations are complicated by the fact that a) many such carers are in paid work, reducing or eliminating their entitlement to the Social Income, which would be a means-tested benefit, b) some carers would not want to be paid, as it could be seen to somehow demean the loving nature of the work, c) some of the care might be for relatively short periods of time and d) there is already a complex ongoing policy discussion about how much carers should be paid, which tends to be at least at the minimum wage for the small number of people who qualify under current schemes.

Especially in light of the arguments for informal care to be paid at some kind of wage rate (that is, much higher than a Social Income would be), in this instance the Social Income might best function as a kind of simple stopgap, available for those who need to take short periods of time out of work to care for family members, and given on receipt of some simple proof like a medical certificate for the relevant relative. More demanding or long-term informal care might need to be covered through some kind of wage-based payment, as above.

Just how many people would be eligible for a (means-tested) Social Income is therefore almost impossible to determine. But assuming that, say, for argument’s sake around 50,000 people were accessing it as informal carers at any one time, the cost would be approximately $1 billion.


This is also hard to estimate. Surveys suggest 28% of New Zealanders volunteer, of which 14% do 25 hours or more. That is roughly 158,000 adults. But of those, perhaps 25% are over 65, and already eligible for New Zealand Super. That would leave roughly 120,000 volunteers, which at $19,000 a year each would cost $2.3 billion. (The economic value of volunteering in New Zealand has been put at $3.3 billion.) The true cost would undoubtedly be less, as many people would not want payment. On the other hand, there would be some behavioural effects, as those in low paid work decided they would rather volunteer and receive a Social Income.

The above estimates have a very large degree of imprecision. So, without attaching too much importance to the figure, one could assume that some of the above effects cancel out and the cost would be roughly $2.3 billion.


The Social Income could be extended to tertiary students, depending on society’s views as to desirability and affordability. (Tertiary students, after all, are already disproportionately drawn from well off families, and likely to earn more over their lifetimes than others do.) The current maximum student allowance is $230 a week, the equivalent of $11,960 a year, and costs taxpayers $581 million.

Increasing the maximum allowance to the Social Income level of $19,000 a year would represent a roughly 40% increase to that cost, taking it to around $800 million – that is, $220 million extra. That assumes the current means-testing for student allowances was extended into the Social Income regime. (Estimating the exact cost under the current means-tested abatement regime would require calculations too sophisticated to be carried out here.)

Overall behavioural effects

With a more generous payment available for a range of activities, the number of people leaving paid work and taking up those activities would undoubtedly increase. It is impossible to know how big that effect would be, and people are less sensitive to financial rewards than is sometimes thought. But one could, for argument’s sake, assume a 10% shift across all categories, and therefore roughly a 10% increase in the total cost of the scheme.


Existing beneficiaries: $2.1 billion

Carers: $1 billion

Volunteers: $2.3 billion

Students: $220 million

Basic total: $5.6 billion

Total including behavioural effects: $6.2 billion

Of course, these costings don’t take into account any potential long-term savings. These could be significant, given that – for instance – child poverty is estimated to cost upwards of $6 billion a year (in increased health and other spending), and this proposal would significantly reduce that poverty. But such savings are impossible to estimate without detailed modelling.

Paying for the Social Income

Extra government spending of around $6 billion would obviously have to be generated. The Tax Working Group has estimated that a capital gains tax would raise $6 billion in the long run, but that would take time to achieve and would not leave any money left over for the many other things on which public action is needed.

Rather than a tax levied when people sell assets, we might consider a straightforward wealth tax, of the kind proposed by the lauded economist Thomas Piketty. He proposes a straightforward wealth levy – that is, people pay an annual charge equivalent to a certain percentage of their wealth. This speaks to the fact that people enjoy the benefits of wealth every year they own assets, not just when they sell them.

It does of course create some administrative complexity, but other countries have managed such taxes without overwhelming difficulty. In addition, solutions have to be found for people who have significant assets but limited income from which to pay such a tax. Generally the answer is that the government allows them to defer payment of the tax until their death, or the disposal of the asset, at which point it is paid out of the sale proceeds.

New Zealand’s household wealth is approximately $1 trillion. A simple 1% levy on that wealth would generate $10 billion, enough to pay for a Social Income and some other priorities.

However, Piketty suggests that such a wealth tax should be progressive – that is, the percentage paid increases with the individual’s wealth – to reflect the fact that people with larger fortunes generally get greater returns on them, and to counteract inequality. We might then imagine a scale in which individuals pay no tax up to the value of the average wealth holding, which is approximately $300,000; 0.5% of the value of wealth up to $1 million; 1% on wealth between $1 million and $10 million; and 2% of wealth over $10 million.

How much this would raise is harder to estimate. But the poorest 80% of New Zealanders, who have only approximately 30% of all assets, would generate very little tax, as most would have less than $300,000 each. (80% of the population can have less than half the average wealth holding because the wealth distribution is heavily skewed towards the top.)

It might nonetheless be possible to generate $10 billion in taxes (the same as a straightforward 1% levy) from the $700 billion in assets held by the wealthiest fifth. The few hundred individuals on the annual Rich List, for example, have $100 billion in wealth, which alone would generate nearly $2 billion (not all of them are resident for tax purposes).

No Comments Share on Twitter Share on Facebook

Say something

Your email address will not be published. Required fields are marked with a grey bar.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>