How to get to a UBI

by John Quiggin on March 20, 2020

Last year I published a book chapter arguing that the first step way to get to a Universal Basic Income in Australia was to expand the existing benefit system, increasing payments and removing conditionality (relevant extract over the fold).

This is often called a Guaranteed Minimum Income (GMI). I counterposed the GMI approach to the alternative of making a small payment to everyone in the community, and then trying to increase it over time. I suggested three initial steps

Assuming a ‘basic first’ approach is preferred, how might it be implemented? Three initial measures might be considered:

(i) increase unemployment benefits, at least to the poverty line;

(ii) replace the job search test for unemployment benefits with a ‘participation’ test;

(iii) fully integrate the tax and welfare systems

We are already on the way to taking these steps. Having floated the idea of a separate benefit for people who lose their jobs due to the virus crisis, the Australian government has quickly abandoned it in favour of an increase in existing benefits. This is supposed to be temporary, and, in theory, at least, there has been no change in compliance efforts like work testing. But ‘temporary’ will turn out to be a long time, and compliance efforts are going to be impossible until things return to normal.

Extract from J. Quiggin, Basic or Universal? Pathways for a Universal Basic Incom, in Implementing a Basic Income in Australia: Pathways forward, Melbourne University Press, Melbourne, 2019.

The alternative is to start with ‘Basic’ rather than ‘Universal’. That is, begin by providing sufficient income to support a decent standard of living to those most in need, then expand it to the entire population. This approach is most naturally associated with a Guaranteed Minimum Income. 

Existing benefit systems potentially fall short of the GMI in three ways. First, each benefit in existing systems, such as that in Australia, is conditional on eligibility requirements such as disability or job search activity. Second, means-tested benefits such as those offered in Australia are subject to clawbacks that imply high effective marginal rates of taxation if people take work along side the support. Finally, with exceptions such as the old age pension the benefits are typically insufficient to lift recipients out of poverty.

Over  recent decades, access to basic incomes has become steadily more difficult in all these respects. The case of unemployment benefits, noted above, is typical . Similar cuts and restrictions have been imposed on disability benefits and supporting parents benefits in Australia and elsewhere in the world. These cuts have been driven by a combination of neoliberal drives to reduce public spending and conservative hostility to welfare recipients, reflected in the use of stigmatizing terms such as Joe Hockey’s distinction between ‘lifters’ and ‘leaners’.

A ‘basic first’ approach would require reversing these trends and would therefore entail immediate and sharp political division between advocates of a basic income and supporters of the push to restrict welfare benefits to the ‘deserving poor’.  Political implications are discussed in more detail in Section 4.

2.3 How to get there

Assuming a ‘basic first’ approach is preferred, how might it be implemented? Three initial measures might be considered:

(i) increase unemployment benefits, at least to the poverty line;

(ii) replace the job search test for unemployment benefits with a ‘participation’ test;

(iii) fully integrate the tax and welfare systems

2.3.1 Increasing unemployment benefits

The basic old age pension in Australia is around 28 per cent of Male Total Average Weekly Earnings (MTAWE) for single pensioners and 42 per cent for couples. This income has proved sufficient to eliminate, almost completely, poverty among the old, who were once the most exposed to privation. The same level applies to Service Pensions and Disability Support Pensions.

By contrast, unemployment benefits (now given the Newspeaky name Newstart) were briefly set equal to old age pensions at 25 per cent of MTAWE under the Whitlam government. However, a long series of cuts and freezes have reduced access to benefits and cut their relative value to around 18 per cent of MTAWE today

2.3.2 Participation income

While social acceptance for a completely unconditional basic income is a long way off, a ‘participation income’ as proposed by Atkinson would have many of the same effects.  The criteria for earning a basic income would no longer be based on market production but on a social assessment of value.  Participation in this context would include full-time study, raising children, and voluntary work.

The creation of a participation income would automatically raise the question ‘what kinds of activities’ are socially valuable. The default assumption, in a market society, is that the social value of any activity is reflected in the market income it generates. This assumption is explicit in neoliberal thinking about public policy, but it is also widely shared in the community in a more or less qualified form. Debate about a participation income would therefore involve fundamental changes to the assumptions underlying the neoliberal economic and social order.

On the one hand, supports of a BI would seek to extend the concept of participation to encompass commitments to artistic, cultural and sporting endeavours, even if these were not at a level sufficient to generate a market income, or to qualify for existing forms of public support, such as arts grants. On the other hand, debate over BI would focus attention on the fact that some activities generate large market incomes but yield little, or even negative, social value. The activities of the financial sector provide an example.

2.3.3 A fully integrated tax-welfare system

As discussed above, a fully implemented basic income would imply an integrated tax-welfare system, in which the distinction between means-testing and taxation would disappear. A step towards this goal would be the inclusion of benefit payments in taxable income, with a corresponding, or larger, reduction in clawback rates.

Completely integrating clawbacks into the tax system would clarify the high effective marginal tax rates currently faced by benefit recipients (commonly above 60 per cent). This would provide a counter-argument to the spurious claims that the marginal rate faced by high-income earners (less than 50 per cent) constitutes an unreasonable disincentive to work effort.



Tim Worstall 03.20.20 at 11:07 am

” increase unemployment benefits, at least to the poverty line”

Which poverty line? 60% of median income?

Which is going to get expensive when we get the insistence that one income should provide a household with 60% of median household income.


ccc 03.20.20 at 1:31 pm

In concert with your proposal the virus crisis is also a great opportunity to seed social wealth funds. Condition any corporate corona bailout on transfer of a healthy chunk of corporate stock into public SWF ownership. As suggested by Matt Bruenig among others.


John Quiggin 03.21.20 at 3:22 am

@Tim “Which is going to get expensive when we get the insistence that one income should provide a household with 60% of median household income.”

Not following this. Can you spell out your point with reference to a two-adult, two-child household

@ccc I’ll be posting soon on the crisis and the case for socialism


Tim Worstall 03.21.20 at 11:28 am

“Can you spell out your point with reference to a two-adult, two-child household”

OK. I’m picking up on something that’s commonly claimed in the US at present, that the one income at minimum wage is “too low” because it doesn’t allow the support of a family of four, two adults, two kids. While that minimum wage does cover – by American and thus low standards – getting one adult up to or around that poverty level.

So, you say that unemployment benefits should get to the poverty line. Fine. But which poverty line? One individual unemployed, benefits are sufficient to get them up to something like the US minimum wage? Or to the more usual poverty line of 60% of median income? Or that unemployment benefits for one should get a four person family up to the poverty line of 60% of median household income?

If you’re suggesting the last then I can see considerable problems ahead. UK median household income is of the order of £25,000 a year. Unemployment benefits for one person out of work of £288 a week (60% of that median household income)? That’s pretty high in relation to the wages that could be earned by going out to work. It’s £7.20 an hour on a 40 hour workweek, which is pretty close to the £8.72 for over 25s min wage, above that for 18 to 20s and 50% higher than that for under 18s and apprentices.

I can see the occasional incentives problem if unemployment benefits for one person should be 60% of household median income.


reason 03.23.20 at 1:25 pm

I absolutely loath the concept of a guaranteed minimum income. Unless it works like a UBI, it destroys incentives for low earners. As basically Tim points out. Wouldn’t it be better to phase in a UBI and phase out unemployment benefits altogether. Why are you so keen on arbitrary bureaucracy. Use environmental taxes as a trigger by introducing them and distributing the proceeds. Higher marginal rates (even involving only environmental taxes and VAT for lower earners) and a UBI are the way to go. If we already had a UBI then the corona crisis would be half as bad.


John Quiggin 03.24.20 at 12:52 am

@reason As I’ve explained quite a few times, when you consider the combined impact of marginal tax rates and welfare clawback rates (that is, in the jargon, the effective marginal rate of taxation), UBI and GMI are equivalent. To be more precise, any UBI scheme can be replicated by a GMI with the same effective marginal tax rates, and vice versa. So, the incentive effects are the same. As regards bureaucracy, since the GMI involves removing a lot of conditionality currently attached to welfare payments, it implies less than we have now.


Collin Street 03.24.20 at 5:57 am

@Tim Worstal: it seems to me that your point could have been made much less abrasively and abtrusely if you’d simply said something along the lines of “poverty-level incomes are set with regard to single-income households but a GMI for a two-adult household would presumably be paid individually”.

Comments on this entry are closed.