Andrew Leigh

How Can We Reduce Inequality?*
‘Just Ideas’ Talk #3

Andrew Leigh
Shadow Assistant Treasurer
Federal Member for Fenner
www.andrewleigh.com

Crawford School of Public Policy
Australian National University
20 April 2017

This talk is dedicated to Tony Atkinson

On Sunday 9 April, a handful of well-heeled passengers boarded an eight-seater Cessna Citation Sovereign business jet.[1] Arriving at Chicago’s smaller Midway airport a few minutes before departure, they were whisked onto the tarmac and slipped into their leather seats. After take-off, the passengers on flight EJA342 tucked into their customised meals, along with wines chosen with help from Wine Spectator’s experts. Less than two hours later – quicker than a commercial flight – the twin-engine jet had touched down in Boston.

On the other side of the city, a doctor from Kentucky was sitting on board a plane at Chicago’s largest airport, O’Hare. The airport is one of the busiest in the world, and more than one quarter of all flights are delayed. After everyone had boarded United flight 3411, a flight attendant announced that four passengers would have to get off, to make room for airline employees. The airline initially offered $800 flight vouchers, then when no-one accepted, they announced that they would pick four passengers at random. Three of the passengers left the plane voluntarily. The doctor, David Dao, refused, saying that he had to see patients in Louisville the next morning. United Airlines staff called aviation security officers, who dragged him off the plane.

The contrast between flights EJA342 and UA3411 symbolises much of what has changed in society over recent decades. Over the past decade, the private jet market has doubled, with increasing numbers of Americans (and Australians, for that matter) being willing to pay up to $8000 per hour to charter their own plane.[2] Meanwhile, the flying experience for most passengers has become worse. Most passengers don’t end up with a broken nose and missing teeth like David Dao. But legroom has shrunk, while surcharges for snacks and baggage have grown.[3]

It is doubtless a good thing that flying to new places has become cheaper and more accessible. But in the process, a gap has opened up in how we get around. At the top, private jets abound and Emirates has introduces in-flight showers for first class passengers on its A380. At the bottom, United Airlines recently announced its ‘Basic Economy’ fare, which gives passengers no choice of seating, and allows them to bring on only one small tote bag. Inequality is everywhere – even in the skies.

* * *

This is the third and final of my ‘Just Ideas’ talks on inequality. The first talk, delivered at the University of Sydney, was focused on why inequality matters. The second talk, hosted by Per Capita, sought to explain the rise in Australian inequality. Today, my focus will be on what we can do to reduce inequality. It seems fitting to do this in Canberra, and I am grateful to the Australian National University’s Crawford School of Public Policy for hosting me.

Reducing inequality is a big call. In his new book, Stanford’s Walter Scheidel looks at inequality over history, and concludes that only four things have made societies substantially more equal: epidemics, pandemics, state collapse, and world war.[4]

You’ll be relieved to know that I don’t advocate any of these as means to a more equal society. But I mention Scheidel’s work because it is a useful reminder that the task is difficult, and that even if we adopt fairer policies, progressives should not expect to see an instant transformation in the level of inequality in our society.

Still, it’s a task that is worth embarking upon. Since I gave the second ‘Just Ideas’ talk, the great inequality scholar Tony Atkinson has died, losing his life to cancer on New Year’s Day. This talk is dedicated to him. Tony’s early work was primarily on measuring inequality, but during his career he became increasingly focused on policy remedies. His final book, Inequality: What Can Be Done?, is jam-packed with suggestions about reducing inequality. I will draw a few ideas from it today, but there is no substitute for reading it yourself.

* * *

Let’s start at the beginning. No, actually, let’s start before the beginning. The womb is an amazing organ, providing developing babies with all the nutrients they need to grow from a single cell to an infant weighing a few kilograms. But even in that environment, we know that development differs. Stress, toxins and micronutrients all affect how a child develops in the womb. Babies born too light (below 2.5 kilograms) or too early (before 37 weeks) are more likely to have health complications, less likely to do well at school, and less likely to earn a decent income.[5] In richer Australian households, only 3 percent of babies are of low birth weight. In poorer households, that figure is 7 percent. Similarly, in affluent households, only 4 percent of babies are born early. In disadvantaged household, that figure is 9 percent.[6]

Reducing pre-birth inequality requires reducing the share of pregnant mothers who smoke, drink, use drugs or are subject to intimate partner violence. One of the most promising interventions seems to be nurse home visits, though as a Cochrane Review pointed out, we still need to learn more about why some home visiting programs work, while others have no effect.[7] The Cochrane Review also points to some surprising findings, such as the fact that financial incentives seem to be particularly effective in helping pregnant women quit smoking.[8] The vital importance of a healthy pregnancy also reinforces the value of paid parental leave, properly funded legal aid for family violence victims, and the ability to visit a doctor without worrying about the cost.

Next, there’s education. Now, you’re no doubt familiar with Australia’s fall in test scores over recent years. From the 1960s to the early-2000s, the literacy and numeracy of 14-15 year olds either flatlined or fell slightly.[9] From 2000 to 2015, the international PISA exams show Australian 15 year-olds going backwards in maths, reading and science, with the drop equivalent to 3-6 months of learning.[10] Most recently, NAPLAN results for grades 3, 5, 7 and 9 show only slight evidence of improvement over the period 2008 to 2016.[11]

However, what you may be less aware of is how we’re doing on educational inequality: the gap between students in the top tenth of performers and students in the bottom tenth of performers. This so-called 90/10 gap is a measure of inequality in the school system, and has been shown to correlate strongly with income inequality.[12] 

Using data from the OECD’s latest PISA exams, I looked at the performance of the top tenth and bottom tenth of test score performers.

Our top tenth of students are very good. Not surprisingly, our best outperform the average student in any country – but they also do pretty well against top performers elsewhere. Take a student from the top tenth in any Australian school and send them on exchange to another advanced nation, and – assuming they know the language – they’ll hold their own with the best students there.

But then there’s the bottom tenth. Australia’s lowest-performing students are well below the average in any advanced country. Indeed, if you wanted to find a country where Australia’s bottom tenth are on par, you’d need to be thinking of nations such as Brazil, Indonesia and the Dominican Republic.

Subtract the bottom tenth score from the top tenth score, and you’ve got a measure of educational inequality. For science and reading, Australia’s level of educational inequality puts us in the top quarter of the advanced world – and this isn’t a ranking you want to top. For maths, our test score inequality puts us in the top third of the advanced world.[13]

When I measured the size of the gap, I couldn’t quite believe what I saw at first. In each of these subjects, the typical student at the 90^th percentile is more than five years of achievement ahead of their counterpart at the 10^th percentile.[14] Remember that this is a grade 9 test. So the typical student at the 90^th percentile is performing at year 12 level, while the typical student at the 10^th percentile performs at year 7 level. They’re a whole high school apart.

The same holds for adult tests of literacy and numeracy. Measured in terms of educational inequality among the whole population, Australia’s test score gap ranks us in the top third of the advanced world.[15] Ask a random group of people to critique an essay, count calories, or calculate mortgage rates, and you’ll find bigger differences in Australia than you would in most advanced nations.

Raising educational attainment for the bottom performers requires delivering needs-based funding. Under the former Labor Government, states and territories received additional federal funding, but only if they contributed an additional $1 for every $2 from the Commonwealth. All jurisdictions committed to national education targets, including a 90 percent year 12 completion rate by 2020, and for Australia to be placed in the top five performing countries by 2025.[16] Among the strategies for achieving these targets were stronger teacher education standards, early intervention for children struggling to read, and a better principal performance framework. As a consequence of the change of government in 2013, schools will lose billions of dollars of needs-based funding. The Coalition’s ‘no strings attached’ approach absolved states of any requirement to increase school funding. National targets still exist in legislation, but are rarely mentioned by the federal government.

In ensuring that resources are spent in the most effective way, we can learn from international examples. America’s What Works Clearinghouse and Britain’s Education Endowment Foundation are building a strong evidence base in those countries that allow state and local education authorities to confidently choose the best interventions. In Australia, we don’t have an equivalent. Moreover, we tend not to use much hard science around evaluating educational interventions. Like medicine in the nineteenth century, the main focus is on expert opinion and case studies, rather than rigorous controlled experiments. For the best-performers, this may not matter much. They likely have home environments with pushy parents, plenty of books and strong social networks. But for the bottom tenth, school resources and how they are spent can mean the difference between success or failure – not just in school but in the job market.

Now, social policy. In terms of its immediate impact on reducing inequality, no part of the Australian system does more to equalise incomes than the transfer system. If you think in terms of Gini coefficients – and, let’s face it, who doesn’t? – then you can think of the Australian transfer system reducing the Gini by around 10 points.[17] That’s roughly the difference in disposable income inequality between the United States and France.[18] Australia’s transfer system does twice as much to reduce inequality as our tax system, which I’ll come to shortly.

The genius of the Australian social safety net is that it is amazing efficient in reducing inequality. Put another dollar into the Australian social safety net, and you reduce inequality by more than you would in any other nation in the world. As Peter Whiteford has pointed out, we spend relatively little on cash transfers – around 9 percent of national income, compared to 10-20 percent in the rest of the advanced world.[19] Frankly, it’s not surprising that we have relatively modest welfare spending, given that government is a smaller share of the economy in Australia than in most advanced countries.

But what marks us out from other countries is that what we spend is highly targeted. In most advanced nations, only around one-quarter of cash transfers are income-tested. In Australia, three-quarters of cash transfers are income-tested.[20] If anything, this trend has accentuated over recent years – just think of the scrapping of the Baby Bonus, and a number of bipartisan changes to family payments.

Further changes to our welfare net should bear in mind three principles: that Australia’s welfare system spends much less than most countries, that Australia’s welfare system is better targeted than any other country, and that Australian inequality is at a post-war high. Mindless cuts to welfare spending will increase inequality. Instead, we should be thinking about how to help working-age adults get the skills and employment opportunities they need. In terms of how to ensure that our social supports are targeted, portable and flexible, there is no better policy document than Jenny Macklin’s thoughtful Growing Together report, released last year.

As it happens, using social policy to reduce inequality is almost precisely the opposite of the suggestion that Australia adopt a ‘universal basic income’.[21] Here’s an illustration of how that might work out. Suppose we got rid of all our current cash transfers and replaced them with a flat-rate universal basic income. Current spending would support a payment of around $6000 per person.[22] Every millionaire and billionaire would be thousands of dollars better off. But every pensioner would be in abject poverty – barely able to buy food, let alone pay their bills. Remember that Australia’s social safety net reduces inequality by 10 Gini points, while a universal basic income – by design – has much less impact on inequality. So scrapping the social safety net in favour of a universal basic income would increase the Gini – potentially making Australia as unequal as Latin America.

Some argue that a universal basic income should be paid for by increasing taxes, rather than by destroying our targeted welfare system. But I’m not sure they’ve considered how big the increase would need to be. Suppose we wanted the universal basic income to be the same amount as the single age pension (currently $23,000, including supplements). That would require an increase in taxes of $17,000 per person, or around 23 percent of GDP. This would make Australia’s tax to GDP ratio among the highest in the world.

So next time someone advocates a universal basic income, ask them how they’d like to pay for it: by making Australia the most unequal country in the world, or by making Australia the most highly-taxed country in the world.

In tax policy, reducing inequality needs to become a more central part of policy design. When I was appointed Shadow Assistant Treasurer, a senior policy economist suggested that I keep close at hand a table of the deadweight costs of different taxes, and judge any tax change based on whether it reduced the economic drag of the tax system. It’s a useful rule of thumb, but it’s not the whole picture. When inequality is high and rising, we also need to think about the distributional implications of tax changes. Frankly, we haven’t always done that well enough. The 1975 Asprey Review, the 1999 Ralph Review and the 2015 Re:Think discussion paper placed equity to one side.[23] Even the excellent 2010 Henry Review didn’t put enough emphasis on inequality.

One example of how this plays out is the upcoming expiry of the Temporary Budget Repair Levy in a little over two months’ time, which will reduce the tax rate on incomes over $180,000 by two percentage points. Viewed through the lens of efficiency, this is nothing to go to the barricades over, given that income tax has a middling deadweight cost. But from an equity perspective, it’s a travesty. The top 1 percent has doubled its share of national income over the past generation. Is it really fair to deliver a tax cut where more than nine-tenths of the benefits will go to adults in the top 1 percent? When payment cuts for low-income households have been permanent, why should tax increases for high-income earners be temporary?

The same goes for capital gains tax and negative gearing. These tax expenditures aren’t merely inefficient – combining to drive down the home ownership rate to its lowest level in sixty years. They’re also inequitable, with half the benefits going to the top tenth of the population.[24] In the case of negative gearing, the typical teacher or nurse gets less than $300 a year, while the typical surgeon or anaesthetist gets over $3000 a year.[25]

Another example of where putting equity into tax policy makes a difference is the transition from stamp duty to land tax. This reform, which has been implemented in the ACT, and is being considered in South Australia, has typically been sold on efficiency grounds. Stamp duty has one of the largest deadweight costs of any tax, while land tax has among the lowest deadweight costs.

But it might be worth also considering the equity implications, by looking at who pays stamp duty in Australia. To see this, I simply pulled up microdata from the Household, Income and Labour Dynamics in Australia survey (HILDA), and asked the question: how wealthy are first homebuyers compared with all home owners? It turns out that first time homebuyers have, on average, about half the wealth of average home owners.[26] Now, admittedly stamp duty is paid by all home buyers, and when you include purchasers of investment properties, the pool of homebuyers is similarly wealthy to the pool of homeowners. But to the extent that it falls on people buying their first home, stamp duty on first homebuyers isn’t just less efficient than land tax, it’s also less equitable. Compared with homeowners, first homebuyers are younger and poorer. In the future, today’s homebuyers will doubtless be older and more affluent – but that’s not much comfort when you’re asking them to pay the tax today.

We also need to bring an inequality lens to competition policy. In my second ‘Just Ideas’ talk, I discussed how work by Tony Atkinson had forced me as Shadow Minister for Competition to stop thinking of competition policy just as being a matter of efficiency, and start thinking about its distributional consequences. We live in a society where some people shop in the supermarket without bothering to look at the prices. Others check every price, then anxiously monitor the total at the cash register – sometimes embarrassedly asking the cashier to put an item aside ‘for later’. For disadvantaged Australians, competition policy isn’t a luxury – it’s a necessity. And yet as Adam Triggs and I found when we analysed nearly 500 Australian industries, most are concentrated markets.[27] When petrol companies engage in tacit collusion to increase prices, the crunch is harshest on those who buy fuel $20 at a time, not those who fill the tank and charge it to a corporate fuel card.

Uncompetitive markets may be the culprit for several worrying economic trends that we have seen in Australia, including more wage inequality, a rising capital share, lower business investment, declining research spending and a drop in new business formation.[28] Recent OECD research has shown a widening productivity gap between ‘frontier firms’ and ‘laggard firms’, which may be related to the market power that big players can deploy.[29] New analysis across US industries finds that rising concentration has led to a fall in the labour share, with ‘superstar’ firms using excessive mark-ups to increase their profits.[30]

Market concentration is a significant challenge, and I do not pretend to have all the answers. But a starting point is to toughen our competition laws. We should increase civil penalties under the Australian Consumer Law from a maximum of $1.1 million to $10 million, bringing consumer penalties in-line with competition penalties. We should adopt the European Union’s penalty system for anti-competitive conduct, which is based on 30 per cent of the annual sales. We should increase the competition watchdog’s litigation budget, and we should give it a market studies function.

But we can also ensure that inequality is within the remit of the competition laws. For example, courts should apply higher penalties for conduct that targets or disproportionately impacts disadvantaged Australians. The competition watchdog should have an explicit legislative mandate to put the most vulnerable first. 

Addressing economic inequality also means addressing the widening social gap in Australia. As work by Bruce Bradbury has shown, the income gaps within our cities have grown larger since the 1990s.[31] Compared with two decades ago, well-paid workers are less likely to live in places with low average incomes. Bradbury sums up his findings with the old joke: what did the rich man say to the poor man? Nothing, they never met.

A major risk of social bifurcation is that policymakers and commentators become disconnected from the reality of lived experience in Australia. Looking out from the boardroom, the parliament or the newsroom, it’s easy to forget that half of all households have disposable incomes below $80,000.[32]

A few years ago, when Fairfax asked its readers to estimate where they stood on the income distribution, two-thirds of respondents thought they were further down than they really were.[33] The problem was particularly acute at the top. Among those whose actual incomes placed them in the top 10 percent, four out of five thought they were poorer. Among those whose actual incomes placed them in the top 1 percent, five in six thought they were poorer.[34] If one thing unites the top 1 percent, it’s thinking that they are in the bottom 99 percent.

With poverty greatest in outer suburbia and regional Australia, it’s easy for those who dwell in the central cites not to notice the extent of deprivation. Moreover, Australians are less likely to be involved in institutions that promote class mixing. Church attendance is down and membership in community organisations such as Scouts, Guides, the RSL and Rotary has dwindled. As they have waned, so too has the chances that people of different backgrounds will rub shoulders with one another.

In my 2010 book Disconnected, I suggested ten strategies for building community, ranging from holding street parties to using technology to build face-to-face interactions. As Shadow Minister for Charities and Not-for-Profits, I have been holding forums with charities to exchange ideas on building membership. Today, I want to suggest another idea: Deliberation Day. Originally proposed by academics Bruce Ackerman and James Fishkin, Deliberation Day is the notion that in election years, we should set aside a civic holiday on which citizens are encouraged to come together and debate the nation’s future.[35] It’s easy to dismiss as a virtuous pipedream, but there has been a recent resurgence in town hall meeting attendance in both Australia and the US: delighting Bill Shorten, and shocking some US Republican congressmen. Even if only one in twenty eligible voters turned out, Deliberation Day would still create a conversation among 800,000 Australians – making it the biggest civic conversation in our history.

On the industrial front, we should never lose track of the role that unions play in reducing inequality. Unions bargain for pay equity within and across workplaces, and fight particularly hard for those at the bottom. That’s why falling union membership explains about one-third of the rise in Australian inequality over the past generation.

Laws that make it harder for unions to access workplaces, bargain collectively, and engage on an equal footing with management are all likely to increase inequality. As I noted in the second ‘Just Ideas’ talk, cutting weekend penalty rates will hurt low-wage workers the most. Unions are one of the most powerful social forces for egalitarianism. If you’ll allow me a sentence with two radical ideas: building bigger unions is one of the best ways to make Australia more equal.

* * *

A single speech cannot hope to span the policy gamut of inequality ideas. What I’ve mentioned here should be considered an appetiser, not a banquet. To the extent possible, I have avoided reiterating my proposals in Battlers and Billionaires, and have restrained myself on some of my perennial favourites, such as randomised trials, income-contingent loans, and the use of behavioural economic tweaks.

Even a list of the things I have missed out will miss plenty out. The inequality-busting work that is afoot on the Labor side of politics includes:

• Bill Shorten’s leadership on expanding the Closing the Gap targets
• Brendan O’Connor’s work on defending weekend penalty rates
• Catherine King’s work on strengthening Medicare and preventive health
• Tanya Plibersek’s work on ensuring that everyone with the ability to win a place at university can take up that opportunity
• Linda Burney’s expose of the robo-debt debacle
• Terri Butler’s push for marriage equality
• Chris Bowen’s pledge to task the Productivity Commission with producing five yearly reports on how we are going as nation in improving equality of opportunity and social mobility[36]
• Wayne Swan’s work with the Chifley Institute’s Inclusive Prosperity Commission, which he co-chairs with Michael Cooney – especially the Commission’s 2016 report Inequality: The Facts and the Future
• Jim Chalmers’ forthcoming book with Mike Quigley on automation and inequality[37]

And if we’re talking about global inequality:

• Penny Wong’s push for a stronger Australian aid program
• Shayne Neumann’s advocacy for a more generous refugee program
• Mark Butler’s work to ensure that Australia meets our climate change goals

For the Labor side of politics, egalitarianism isn’t an abstract idea. Reducing inequality is a principle that undergirds how we would govern.

* * *

In 1982, Opposition Leader Bill Hayden was speaking to a business lunch when a member of the audience asked him his view on imposing a tax on robots. Hayden swiftly responded ‘we are not proposing to tax robots at the moment’.[38] But 35 years later, in a February 2017 interview with Quartz magazine, Bill Gates proposed just that: a robot tax.[39]

Gates clearly cares about inequality, and his foundation has done an enormous amount to address global poverty. But a robot tax would be a terrible way of addressing inequality. Surgical robots reduce medical complications. Industrial robots make safer products.[40] Driverless cars may soon start to cut the road toll. As Larry Summers points out, Gates makes the same error as those who advocate tariffs: ‘Gates’s robot tax risks essentially being protectionism against progress’.[41]

The fact that a brilliant thinker like Bill Gates can go astray on the question of how to reduce inequality reminds us of how hard the task is. And yet, as Tony Atkinson reminded us, it need not be beyond our wit to take it on. As Atkinson writes:[42]

In this book, I set out concrete policy proposals that could, I believe, bring about a genuine shift in the distribution of income towards less inequality. Drawing on the lessons of history, and taking a fresh look – through distributional eyes – at the underlying economics, I seek to show what could be done now to reduce the extent of inequality. I do so in a spirit of optimism. The world faces great problems, but collectively we are not helpless in the face of forces outside our control. The future is very much in our hands.

May Tony’s energy, ideas and optimism inspire us to work for a world that is not just more affluent, but also fairer, more ethical and more equal.

Notes