… contains the seed of a good idea but his version is an inequitable mess
That’s the headline for my latest piece in the Guardian , over the fold
It is always disappointing when an important, and potentially transformational, policy idea is introduced in a form that almost certainly guarantees its rejection. That’s even more true in the context of an election campaign, where reasoned analysis of policy invariably takes a back seat to partisan polemics.
Dominic Perrottet’s announcement of a “future fund” in News South Wales to help young people finance education or the purchase of a home falls into this category. In the most attractive presentation, young people could start adult life with a fund of $50,000, enough to pay for a university degree, franchise a small business or, even, with other government assistance, pay the deposit on a modest apartment. Unfortunately, as we will see, the reality is rather different.
Perrottet’s proposal is a drastically cut-down version of an idea put forward in the US by Bruce Ackerman and Anne Alstott in the 1990s under the name “The Stakeholder Society”. Ackerman and Alstott proposed a grant of $US80,000 to all young adults. Allowing for inflation and currency conversion, this would be something like $A250,000 today. The proposal would be financed by taxes on wealth, and by treating inheritances and gifts from close family members as taxable income.
Such a proposal would convert the rhetoric of equal opportunity into something like reality. As always, the children of wealthy and well-educated parents would have a headstart, including better opportunities for school education and access to financial support. But gift taxes would reduce the capacity of the bank of Mum and Dad to transmit inequality from one generation to the next. Part of any intergenerational transfer would be taxed to provide a start for those young people not blessed with wealthy and generous parents.
Although it would be a once-in-a-lifetime transfer, a stakeholder grant would share many of the features of a universal basic income. There would be no means test, and no adverse incentive effects. The funds could be used for a range of reasonable purposes, or paid out as a cash grant if not used by, say, age 30.
Given that about 300,000 Australians turn 18 each year, a scheme with a grant of $100,000 would cost around $30bn a year, comparable to the revenue foregone with the stage 3 tax cuts. A grant on the scale proposed by Ackerman and Alstott would cost around $75bn, implying the need to raise lots of money through wealth and inheritance taxes. Clearly, only the commonwealth government could raise the necessary revenue.
How does Perrottet’s proposal compare to a real stakeholder grant? $50,000 sounds good, but the vast bulk would come from parental contributions. The government’s contribution of $400 a year would amount to a maximum of $7,200 (plus interest earnings) which would not be reached until today’s babies turn 18 in 2041.
Even so, the proposal might be worthwhile if the government contribution was the same for everyone. But under Perrottet’s plan, governments would match parental contributions. So, if parents can’t afford to contribute, their children get nothing. Perrottet softens the edge of this by saying that children whose parents receive family tax benefit will get a government contribution of $200, half the amount for well-off families who can afford the full contribution. That ratio is a pretty clear indication of Perrottet’s relative concern for low-income and high-income families.
How many people would get the full grant? At first sight, $400 a year might not sound like too much for the average family. But remember that this is $400 a year for every child in the family. Not many families are as large Perrottet’s own (he is one of 12 children and has seven of his own). Equally, however, very few large families have the resources available to the Perrottets. For a family with four children, even on a decent income, finding the resources to lock away $1,600 a year could prove very challenging.
Perrottet’s proposal is an inequitable mess, unlikely ever to be implemented. But that should not be the end of the story. Inequality is growing and it is increasingly being locked in over generations. A substantial stakeholder grant to young people could change it.
John Quiggin 
I am reminded of things like “fallacy of composition”, “external costs”, “Goodhart’s Law”, and the like. How would issues of that nature be handled or headed off?
“Gentlemen, you can’t fight in here. This is the war room.” – From the movie Dr. Strangelove.
“Gentlemen. You can’t spend the future fund money. That’s for the future!” – general statement of all neoliberal economists.
Important note, investment spending occurs *in* the present *for* the future.
According to the smh, Premier Perrottet announced yesterday he will not apply for the future fund for his 5 (out of 7) eligible children. His stated reason was an account has already been set up for each of his children. In short, his family does not need tax payer funds (debt, to be more precise).
His proposal allocates public debt (allegedly to be repaid by future generations) of up to $400 p.a.. dependent on the contribution of the parents (or grand parents) up to a limit of about $1000 p.a. except for the initial $400 and $200 for families that are family support. Obviously, it potentially benefits most those, like the Premier, who do not need it.
If the NSW State financial situation wouldn’t be so deep in the red and if the income as well as the wealth distribution wouldn’t be as concentrated as what they are, then IMHO, I’d say who cares whether or not this proposal is implemented. After all there are some parents, who do have enough financial means to provide a little nest egg for their children to ease them into adulthood, but who are too mean to do so (although it is a rather huge nest egg in the case of Gina Rinehardt’s family, it does illustrate my point). So, perhaps only 1 in 100 or 1 in 1000 children will benefit in a non-trivial manner and this is good enough.
Given the actual circumstances, I’d say not a cent of public debt should be incurred that even potentially increases income and wealth disparities. Therefore legislate what Perrottet says he does voluntarily – do adopt his proposal.
As for Ackerman and Alstott’s Stakeholder Society idea, it mimics the ‘initial wealth distribution’ assumption in general equilibrium models. IMHO, if real life would take place in a one period economy (ie complete forward markets for commodities in the Arrow-Debreu model) then it would be a reasonable proposal before privatisation and liberalisation of the financial sector to approximate the initial wealth distribution assumption in these models.
IMHO, attempts to treat GE models as models of a society to be aspired to is not helpful. From my perspective, the usefulness of these models lies in pointing toward important parameters in an economy where all agents accept that individual freedoms and individual responsibility for their lives are important but so is solidarity because risk cannot be diversified to such an extent to base laws on what is philosophically desirable. Solidarity find practical expression via institutions, starting with a government. In such a society, which I assume is not categorically different from contemporary Australia, these GE models say, in the sense of providing an insight, you have to watch the income and wealth distribution, don’t let these distributions become to concentrated. Economic policies should be directed to this objective. (And this objective assists also in implementing policies directed at financial system stability and environmental sustainability because any price implications of such policies do not treat people severely differently.)
Given the actual circumstances, I’d say not a cent of public debt should be incurred that even potentially increases income and wealth disparities via possibly well intended band aids.
EG – “don’t let these distributions become to concentrated.”.
As we now have ML / NN / AI well on the ‘tools we can use ‘ path, can’t we just go beyond GE anything?
My intuition tells me “this theory gives us tools to model uncertainty with dropout NNs – extracting information from existing models that has been thrown away so far.”.
Showing missing parameters, outliers and error of policy vs reality? Above my pay grade, yet I’d like to know.
“Do deep models really need to perfectly emulate humans in every way we can conceive of for these approaches to be considered successful?”
(See Frisron, Hinton and…)
“In this paper we develop a new theoretical framework casting dropout training in deep neural networks (NNs) as approximate Bayesian inference in deep Gaussian processes. A direct result of this theory gives us tools to model uncertainty with dropout NNs – extracting information from existing models that has been thrown away so far.”
…
“Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning”
Yarin Gal, Zoubin Ghahramani
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:1050-1059, 2016.
https://proceedings.mlr.press/v48/gal16.html
Click to access Dropout_as_a_Bayesian_approximation.pdf
Via (a very long proposal re developing a mamalian style AI-Immune Systems – I’m sympathetic. ymmv)
“Then in 2015, computer scientists Yarin Gal and Zoubin Ghahramani realized that using dropout, or turning off various subnetworks within a neural network, and watching how predictions varied, was itself an accurate approximation of Bayesian uncertainty for the entire network. Dropout had long been used to make non-Bayesian deep learners more accurate. It was one technique Alex Khrizhevsky used to train AlexNet in 2012 (see Post 2).
“One thing dropout does, in training, is to make each subnetwork more interchangeable, and the whole network more amorphous and bottom-up rather than hierarchical. Their paper, “Dropout as a Bayesian Approximation”, 2016, has been cited an incredible 18,500 times to date since publication. Just as backpropagation solved the problem oftraining hidden layers in a neural network, Bayesian dropout was needed to make Bayesian networks usable in practice. Researchers now think the range of predictions these systems give under dropout variations in any repetitive context, their output uncertainties, is the best measure of the AIs actual uncertainty. This seems a very important advance, as it gives our AIs useful insight into their own predictive uncertainty.”
From:
“Natural Security: Defensive and Sustainable IntelligenceBuilding Strong and Safe Machines, and Secure Human-Machine Teams. – Part 4 of a Limited Series by John Smart and Nakul Gupta
https://naturalalignment.substack.com/p/natural-security-defensive-and-sustainable
*
Or… 3 ensemble baysian going beyond JQ’s RDEU, and Kahneman & Tversky.
Journal of Behavioral and Experimental Finance
Volume 26, June 2020, 100293
“Fast and Frugal heuristics augmented: When machine learning quantifies Bayesian uncertainty”
https://www.sciencedirect.com/science/article/abs/pii/S2214635019302357
*
Or?
“Ex ante and ex post equilibrium supply curves”
Flavio Menezes & John Quiggin
https://ideas.repec.org/p/qld/uq2004/649.html
It occurs to me that we (almost all who’ve blogged here over the years) have been calling for wiser, non-neoliberal social and economic policies for over two decades. Some have been calling for better policies for much longer than that, up to something like four decades. Nothing lasting has ever happened in that direction in this time span. A few things happened which are now being savagely cut and rolled back like Medicare.
Neoliberalism has only intensified/re-intensified and is still intensifying in my estimate. Trends that can’t continue don’t continue of course. But what breaks first? Where will our society break and crumble first? It very likely will on this policy course. Alternatively, where and how will the neoliberal hegemony break? One of these two at least must logically happen. Of course, both could break down too, sequentially or simultaneously, leading to a highly chaotic situation.
Crises of catastrophic proportions will ensue. The sooner we change course the more chances we will have of saving something, saving ourselves and our fellow citizens. The change of course has to be at least back to mixed economy principles with a strong role for socialist and democratic principles in governance . Socialism needs to be fully implemented/re-implemented in health, welfare, education, utilities (power, water, transport, communications), national resource management, natural monopoly cases and in any other economically strategic arenas not mentioned in that list. Tighter regulation and compliance will be needed in all arenas especially finance, environmental issues and public health issues.
Without all of these changes we, or a very significant proportion of us (likely 20% to 80% of the total population) are doomed to poverty, illness and premature death. Life will be nasty, brutish and short. This will apply even to a large proportion, perhaps even to a majority, of the current middle class. The middle class, even the upper middle class, should not suffer under any illusions that they are safe under runaway climate change, endless COVID-19 and widespread social breakdown. They are not.