Image by Conor Lawless CC 2.0 via Flickr

This post is the first of a four part series on the topic.

The political and media storm about “zero net taxpayers”

Following Duncan Storrar’s appearance on ABC’s Q&A on 9 May 2016, one of the milder reactions to his questions about tax cuts was an article in The Australian“ABC’s ‘budget fairness’ victim pays no net tax”.  The article pointed out that Mr Storrar was “sporadically working as a truck driver on $16 an hour and ­relying on a $520-a-fortnight ­Austudy ­allowance to survive”. As a result he was labelled as a  “zero net taxpayer”, an expression used to describe people who either pay no tax or whose benefits received from government  are greater than the taxes they pay.

This had followed an opinion piece in The Australian on 7 May by Peter Van Onselen which asked

“How is it fair that half of all workers pay no net income tax?”  An article by Maurice Newman in The Australian on 29 April similarly argued “According to Australian National University researcher Ben Phillips, only 43 % of the adult population excluding public sector workers are net taxpayers, meaning more than 50 % of voters rely fully on political patronage for their income.” (The meaning of “relying on political patronage” is not spelled out.)

There have in fact been many stories in The Australian about “zero net taxpayers” including  “one in two voters is fully reliant on public purse”  in April 2016, “3.6 million households pay no net tax after churn” in February 2016, “a leaner world as households make the most of welfare benefits” in December 2015, and “welcome to the welfare nation: half of  Australia’s families pay no net tax” in May 2014 (in The Daily Telegraph).

The Australian’s Adam Creighton had made a similar argument in March 2014 under the headline “No, the Rich Don’t Pay a ‘Fair Share’ of Tax. They Pay All of It.” According to ABS figures which included cash benefits plus other government health, education and welfare services as well as indirect taxes, he wrote, “only the top fifth of households ranked by their income – those with incomes of more than $200,000 a year in the financial year ending June 2012 – pay anything into the system net of the value of social security in cash and kind received.”

A related argument was made by then Treasurer, Joe Hockey, in his speech to the Sydney Institute following the 2014 Budget : “Whilst income tax is by far our largest form of revenue, just ten % of the population pays nearly two-thirds of all income tax.  In fact, just 2 % of taxpayers pay more than a quarter of all income tax. Maybe these taxpayers would argue that the tax system is already unfair.”

Most recently, in an address to Bloomberg on 25 August this year, the Treasurer Scott Morrison has argued

“A generation has grown up in an environment where receiving payments from the Government is not seen as the reserve of the disadvantaged, but a common and expected component of their income over their entire life cycle, and not inconsistent with self-reliance. On current settings, more Australians today are likely to go through their entire lives without ever paying tax than for generations and more Australians are likely today to be net beneficiaries of the Government than contributors — never paying more tax than they receive in government payments. There is a new divide — the taxed and the taxed-nots.”

It is clear that the “taxed” and the “taxed-nots” refer to the same concept as taxpayers and “zero net taxpayers”.  This address was followed the following day by an editorial in The Australian “Indebted Australia cannot afford entitlement culture” which argued “Soon, less than 50 % of households will be net taxpayers, meaning they pay more income tax than they get in welfare payments.  Overly generous family tax benefits and childcare subsidies are just two examples of the problem at hand.”

A related concern was expressed in an opinion piece in The Age in June 2016 criticising the government’s proposed changes in superannuation tax concessions: “If fewer and fewer people pay a continually growing share of taxes, they will rebel. As the ratio of net beneficiaries who get more in benefits than pay in taxes rises, citizens will divide into those who work and others who vote for a living. Clawing back the dependency state will become increasingly difficult.”

Apparently, the United States (US), New Zealand  (NZ) and the United Kingdom (UK) also have zero net taxpayers. The best known expression in the US is Mitt Romney’s comments in the 2012 US Presidential campaign that 47 % of Americans paid no income tax (and would therefore be more likely to vote for the other side). It turns out that Donald Trump had cited similar figures in 2011, which were based on an earlier report by the Heritage Foundation, the 2010 Index of Dependence on Government, which had noted:

“The more people there are that depend on government subsidies for much or all of their existence, the smaller the percentage of tax-paying citizens who finance those subsidies. Combined with Congress’s loss of control over the national budget, an impending tipping point is becoming increasingly likely. The Tax Foundation reported recently that the percentage of tax returns filed that required no taxes to be paid in 2008 passed 35 percent.[2] Even more of a wake-up call, the Tax Policy Center says that the percentage of all taxpaying units—single, married filing jointly, head of household, married filing separately—who paid no taxes in 2009 is close to 47 percent.[3]

In New Zealand, a media report in June 2016 noted that “More than one in four households are contributing nothing to New Zealand’s tax take. … 663,000 households – or 40 % – receive more in tax credits and other benefits than they pay in tax. Thousands more are neutral contributors, or are close to it. Households earning less than $50,000 receive more in credits than they pay in direct income. On a net tax basis, 48,000 households pay 28 % of all tax. By comparison, the top 3 % of individual income earners, earning more than $150,000 a year, pay 24 % of all tax received.”

The same theme has emerged in the UK.  In 2014, the Daily Telegraph stated that “the highest paid 3,000 people in the UK pay more income tax than the bottom nine million, according to official Government statistics.  More recently, the same newspaper and the Guardian  reported that the Institute for Fiscal Studies estimated that the proportion of working-age British adults who do not pay income tax has risen from 34.3 % to 43.8 %, equivalent to 23 million people.

Most strikingly, the “zero net taxpayer” theme has been used in explaining the UK’s recent referendum decision to leave the European Union.  Writing in the 28 July 2016 edition of the London Review of Books on Brexit Blues, novelist John Lanchester has noted

“If I had to pick a single fact which has played no role in political discourse but which sums up the current position of the UK, it would be that most people in the UK receive more from the state, in direct cash transfers and in benefits such as health and education, than they contribute to it. The numbers are eerily similar to the referendum outcome: 48 % net contributors, 52 % net recipients. It’s a system bitterly resented both by the beneficiaries and by the suppliers of the largesse.

What taxes?  What benefits?

What do these figures mean?  On the surface at least the suggestion that half, or more, of the Australian population are “zero net taxpayers” sounds surprising if not alarming.  How can it possibly be sustainable that a majority of people receive more in benefits than they pay in taxes?  Moreover, is the number of people who receive more in benefits than they pay in taxes increasing, and if so why?

The media and political debate about who is a “zero net taxpayer” refers to three different measures of taxes paid to government and benefits received from government by individuals or households.

1. Is an individual or a household paying any income tax? (What they get for their taxes is ignored.)

2. Is an individual or a household paying more in income tax than they are receiving in cash benefits, social security or welfare payments (technically called “transfers”) from the government?

3. How do the total direct and indirect taxes of all kinds paid by an individual or household compare with the total cash benefits and public (social) services they receive from government?

Each of these measures is based on a different definition of a “taxpayer” and the last two include different measures of benefits. Each of these measures also draws on different statistical sources.  This can result in quite different figures being quoted.

The majority of the media articles and comments on “zero net taxpayers” refer to the second of these measures.  They look at the combination of income taxes paid and social security cash benefits received and calculate the extent to which one side of the ledger offsets the other. This also appears to be what the Treasurer was referring to in his 25 August Bloomberg address.

1. Who pays income tax?

A recent report by the Grattan Institute provides estimates of trends in the share of people paying income tax by age group since 1999-2000, using administrative statistics from the Australian Tax Office. This is the first concept referred to above and the one used by Mitt Romney in the 2012 US election.

Table 1 shows that in 1999-2000 it was estimated that 59% of individuals aged 18 years and over paid income tax, while by 2013-14, that share had declined to 54% – meaning that 46% were not taxpayers.  For people of working age, the decline in the share of taxpayers was from 66% to 62% – although it had been as high as 70% in 2007-08, just before the Global Financial Crisis.  The share of people aged 65 and over paying income tax is much lower and fell from 25% in 1999-2000 to 16% in 2013-14.

Table 1: Taxpayers by Age (% of age group), Australia, 2000 to 2014

Fiscal year ending Over 65 18-64 Over 18
2000 24% 66% 59%
2001 15% 66% 58%
2002 16% 66% 57%
2003 17% 66% 58%
2004 18% 67% 59%
2005 20% 67% 59%
2006 21% 69% 60%
2007 20% 67% 59%
2008 16% 70% 61%
2009 14% 64% 55%
2010 13% 62% 54%
2011 15% 63% 55%
2012 16% 65% 56%
2013 16% 62% 53%
2014 16% 62% 54%

Source: Daley, J., Coates, B., Young, W. & Parsonage, H., 2016, A fairer super system: assessing the 2016 tax reforms, Grattan Institute.

The Grattan Institute argues that these trends are mainly the result of deliberate policy changes, including the introduction of the Senior Australians and Pensioners Tax Offset (SAPTO) in the early 2000s which saw a very large fall in the share of taxpayers in this age group (from 24% to 15%); and the decision by the Howard government in 2007 to abolish income tax on superannuation withdrawals for those aged 60 years and over, as well as the general income tax cuts in 2008.

Many Australians face lower taxes when they reach age 65 – they pay less in income tax than a person with the same total income under 65 years of age. Using further data provided by the Grattan Institute[1] it can be calculated that the introduction of the SAPTO was associated with a fall in the number of taxpayers aged 65 and over by nearly 200,000. The superannuation tax changes and subsequent income tax cuts in 2007 and 2008, saw a fall of nearly 140,000 taxpayers in this age group. Overall there were slightly fewer taxpayers aged 65 and over in 2013-14 than in 1999-2000, even though the population in this age group had grown by more than one million over the period.

Put another way, people over the age of 65 increased from 16.5% to 18.9% of the adult population over this period, but accounted for more than one third of the number of people not paying income tax in 2013-14 and 41% of the increase since 1999-2000.

In addition to the effects of policy changes, this also reflects trends in the labour market and in wages. The fall in the number of working age taxpayers from 70% in 2008 to 64% in 2009 is partly due to income tax cuts and partly the result of the Global Financial Crisis (GFC).  Since 2008, ABS data show that 60% of the additional jobs in Australia have been part-time, many of whom are likely to be below the tax free threshold of $18,200. In addition, as the Reserve Bank  has noted, Australia is currently experiencing the longest period of low wage growth since the early 1990s recession, with a significant impact on income tax collections.

2. Income tax and cash benefits

The second definition of a “zero net taxpayer” is broader. It involves comparing the income tax payments of households or families and then subtracting the social security cash benefits they receive from government in the form of pensions, unemployment payments, family benefits and other forms of assistance.

The assumption underlying this approach involves treating cash benefits or social security payments as “negative taxes”. (A related concept is the suggestion by Milton Friedman that welfare payments could actually be replaced by a Negative Income Tax.)

This calculation requires a survey of household incomes rather than the administrative statistics on individual income taxpayers because cash benefits or transfers are provided, and means tested, on a family unit basis in many cases. Household income surveys generally collect information on both taxes paid and benefits received, whereas the relevant administrative data from the Tax Office and the Department of Social Services collect either one or the other.

Given that around 54% of adult individuals pay positive income tax (as indicated above), we would expect that netting off cash benefits should reduce this share further.  The extent to which this occurs, however, will be moderated by the fact that for many of the people not paying income taxes this will be because they have low incomes and already rely on cash benefits for their main income source. In addition, if two or more individual non-taxpayers live in the same household – for example an age pensioner couple or an unemployed couple – then there will be a difference between the share of individuals and the share of households who are not taxpayers.

3. Direct and indirect taxes and government spending

Income tax is not the only tax. There is a wide range of other direct and indirect taxes levied by the federal government, state and territory governments and local councils.

As Duncan Storrar commented  when it was pointed out that he was unlikely to benefit from income tax cuts because of his low income, “I pay tax every time I go to the supermarket. Every time I hop in my car.” Mr Storrar is well aware that he bears the incidence of Goods and Services Tax (GST) on a wide range of goods and services, such as shaving cream and soft drinks, books and newspapers, and payments to his mechanic or barber. Mr Storrar would pay fuel excise on petrol for his car. These indirect taxes are passed on to the consumer through higher prices.

Similarly, in a “Factcheck” on Mitt Romney’s 47% figure, US business magazine Forbes pointed out that “because all working Americans are obligated to pay payroll taxes—the money that supports Social Security and Medicare automatically withdrawn from our paychecks—the actual number of households in America paying nothing was closer to 17 % in 2009 and, more typically, 14 % in non-recessionary years.” Similar points have been made by the Center on Budget and Policy Priorities.

This brings us to the third approach to defining zero net taxpayers, which is broader again. This approach involves modelling the impact of indirect taxes such as the GST and excises on petrol, alcohol and tobacco, or social security tax in other countries, together with direct taxes such as income tax and land tax. This approach then balances this more comprehensive definition of taxation with a more comprehensive definition of spending, to include government funded or provided services such as health, education, and services such as child care, public housing and disability care and aged care.

The UK figures quoted by John Lanchester above are based on research by the UK Office of National Statistics  (ONS) which analysed the effects of taxes and benefits on household income for the financial year ending in 2015. The figures cited by Adam Creighton in The Australian are derived from the Australian Bureau of Statistics (ABS) which every five years or so undertakes a similar study, in this case for 2009-10.

The ONS Study and the ABS Study analyse the redistributive effects on households of direct and indirect taxation and benefits received in cash or kind. The methodology used in each of these studies is virtually identical; indeed, the ABS study is explicitly based on the UK methodology (with some minor differences).  These studies combine household income surveys collecting information on direct taxes paid and benefits received with modelling of the impact of indirect taxes and government services. [2]

For example, detailed data on the expenditure patterns of households is used to estimate how much of their spending is subject to the GST, or if they are smokers how much they pay in tobacco taxes, and so on.  The value of government provided or subsidised health care is valued as the cost of an insurance policy that would provide this level of services, according to the age and sex of household members. Education spending and child care spending are allocated to families with children according to the ages of their children.

However, even the ABS studies of the impact of government taxes and benefits  do not include all taxes and all government spending.  As the ABS points out:  “The aim of the study has been to allocate only those benefits and taxes relatable to particular types of households.  No attempt has been made to allocate the whole of government expenditure and revenue” (Paragraph 78).

For example, the ABS do not allocate government spending related to public debt, or on general public services, fuel and energy, transport and communication, public order and safety, defence, recreation and culture, agriculture, forestry and fishing, mining and construction and other.  The ABS do not give precise numbers for this non-included spending, but based on the chart in Paragraph 78 of the ABS publication, there may be as much as $200 billion in spending in these other areas  in 2009-10 that is not included. This can be compared to around $234 billion in spending on health and social security and welfare in that year, implying that these studies leave out nearly half of all government spending.

Moreover, while the ABS allocate about 91 % of income taxes to households, they allocate less than two thirds of taxes on production, a gap of around $47 billion in 2009-10. They also do not include about $10 billion in income taxes (mainly capital gains taxes), plus another $50 billion in other taxes.

What would happen if we added these other expenditures and taxes into the picture? In the case of Australia, we do not know, but Edward Wolff – a prominent US researcher in this field – and colleagues undertook such a study for the United States (Wolff et al., 2007). They allocated items of “public consumption” including police, fire services, law courts, prisons, highways, public transport, and airports and so on to a mix of the household and non-household sector. Applying a range of distributional assumptions,   Wolff et al. (2007) estimated that US government spending on social security and healthcare strongly favoured lower income households. However, the distribution of public consumption services while progressive as a percentage of “comprehensive income” was relatively less favourable – with the dollar value of spending being twice as high for the richest 10 % of the population as for the poorest 10 %.

In sum, all the figures used to estimate whether people are “zero net taxpayers” are incomplete in one way or another.

Part two of this series will discuss net tax thresholds.

The full article on which this post is based can be found at the TTPI website.

[1] Daley and Parsonage, personal communication, 2016

[2] There is an important difference between the figures quoted by John Lanchester and those quoted by Adam Creighton. The figures for Great Britain are directly calculated by the ONS from microdata for individual households, while the figures for Australia are derived from published tables from the ABS which summarise the distribution of income components by quintiles (20%) of household income, with the estimate of “zero net taxpayers” being the income quintile where the average total taxes paid by the quintile start to exceed the average total benefits received by the quintile.  Because this compares averages within the quintile, it is less precise than the figures for Great Britain.

I am grateful to Miranda Stewart, John Daley and Ben Phillips for comments on earlier drafts of this Policy Brief.  Any errors are my own.

Leave a comment

Your email address will not be published. Required fields are marked *