Andrew Leigh

Why should we care about inequality?

The 2015 Economic and Social Policy Public Lecture

University of Wollongong

Dutch economist Jan Pen once suggested a simple way of visualising the extent of inequality in a society. Imagine, he suggested, a parade, in which each person’s resources were represented by their height.

Suppose we were to conduct such a parade in Australia. People of average wealth would be average height. Those with half the average wealth would be half the average height. Those with twice the average wealth would be twice the average height.

Let’s suppose the parade took an hour to pass you. What would you see?

For the first half a minute, people would be literally underground. These are the people with more debts than assets. Perhaps they are homeless, but have credit card debts. Or they are a business owner about to go bankrupt.

Then would come the little people. For the first few minutes, they are no bigger than lego figures. They might have some clothing and a television, but little else. By the ten minute mark, people are the size of a child’s doll. They might own an old car.

Twenty minutes have gone by, but still the marchers are no taller than a newborn baby. Most probably don’t have regular work – they might have odd jobs, or be reliant on the pension. Few would dream about them – or their children – breaking into the central Sydney property market.

It’s now half an hour. The watchers are getting bored. Heights aren’t rising much. Are we really only halfway through the parade?

They notice something else, too. Even although the parade is only halfway over, the people are still short. They’re about 100 centimetres tall; about the height of a four year-old. Thinking about it, one watcher realises that this is because wealth is so skewed – the mean is much higher than the median.

Forty-five minutes in, and the watcher can now look the marchers in the eye. These are homeowners with well-paying jobs – in many cases probably with two full-time workers in the household.

Ten minutes to go, and the parade is suddenly getting interesting. Now, the marchers are two and a half metres tall, and the heights are rising fast. These people own multiple homes, and most likely have earnings well above average. They’ve done well over the past generation, with earnings having risen three times as fast for the top tenth of Australian workers as the bottom tenth.

Five minutes till the end of the parade, and the marchers are four metres high. Now come the giants. Ten metres high, then fifty, then one hundred metres high. Their shoes are now as big as the watchers – their faces as high as office buildings.

One-thirtieth of a second before the end of the march, and we’re into the BRW rich list. The poorest person on the BRW rich list is twice as high as Centrepoint tower. The rest are taller still. Now, their heads poke into the clouds. The tallest person in the parade is nearly 50 kilometres high – halfway to outer space.

Does inequality matter?

But does this matter? Some economists have argued that we shouldn’t worry about inequality. According to Robert Lucas, ‘of the tendencies that are harmful to sound economics, the most seductive, and in my opinion the most poisonous, is to focus on questions of distribution’. Martin Feldstein criticises a focus on inequality as being mere envy, and writes that ‘there is nothing wrong with an increase in wellbeing of the wealthy or with an increase in inequality that results from a rise in high incomes’.[2]

Yet economics has a long history of being concerned about not just the average, but also how it is distributed.

Adam Smith wrote that ‘This disposition to admire, and almost to worship, the rich and the powerful, and to despise, or, at least, to neglect persons of poor and mean condition… is… the great and most universal cause of the corruption of our moral sentiments’.[3] The economics Nobel has been awarded to several economists who worked on inequality, including Simon Kuznets, Amartya Sen, Joseph Stiglitz and Paul Krugman. The Clark Medal, the young person’s Nobel Prize, recently went to Emmanuel Saez, one of the world’s leading inequality scholars. According to Google Books, the share of books that discuss inequality has doubled since 1960, and is now at an all-time high.[4]

There are two sets of arguments typically used against inequality: intrinsic and instrumental. Intrinsic arguments are that inequality has a cost in and of itself. Instrumental arguments are that we should care about inequality because it makes us sick, increases crime, or reduces the savings rate.

In my view, economists – and many others who care about inequality – have been too quick to skip over the intrinsic arguments in favour of instrumental ones. In so doing, we’ve missed out on the chance to engage about what inequality really means, to talk about the ethics of inequality. Because the instrumental arguments are fragile, we’ve effectively put our worst foot forward.

Intrinsic concerns about inequality

Let’s start with the intrinsic arguments. Imagine two societies, both with the same average income. In one of them – call it Fairfield – the resources are pretty evenly distributed. Some people have more than others, but the best-paid person is only paid ten times what the worst-paid receives. In the other society – call it Unfairfield – one person has all the money, and everyone else has nothing.

Where is average happiness higher? If you’re like most people, you’d say that Fairfield is a happier place than Unfairfield. While it’s true that interpersonal comparisons of utility are methodologically tricky, it still strains credulity to imagine that one person’s gain from having all the money outweighs everyone else’s loss from having nothing. Most of us intuitively believe that a dollar brings more happiness to a homeless person than to a billionaire.

The philosopher John Rawls formalised this idea in his famous thought experiment about the ‘veil of ignorance’.[5] Imagine, he said, that you were waiting to be born. From inside your mother’s uterus, you would be ignorant as to whether you were about to be born tall or short, black or white, rich or poor. Given that, would you prefer to be born into Fairfield or Unfairfield?

Rawls’ answer helps us think not only about inequality, but also about redistribution. From behind the veil of ignorance, he argues, we would want inequality only so far as it raises the wellbeing of the most vulnerable. You probably wouldn’t want a perfectly equal society, where rewards are divorced from effort, since that model has been shown to work badly. A bit of inequality boosts innovation, and encourages entrepreneurship.

Moreover, as economists know, redistribution comes at a cost. Employment taxes reduce work effort – what Arthur Okun called the problem of the ‘leaky bucket’.[6] A good tax-transfer system should bear this in mind in deciding how much it should do about reducing inequality.

Some argue that policy should ignore ‘equality of outcome’, and focus only on ‘equality of opportunity’. This sounds good on first blush, but soon runs into practical problems. Are we really prepared to give a poor child all the resources available to the most affluent? Even if we wanted to do so, could public policy ever really make up for differences in the home environment?

Moreover, as inequality scholar Tony Atkinson points out, a system that cares only about what happens until the starting gun is fired is effectively a system that says we shouldn’t care if one of the runners trips over.[7] From an ethical perspective, why should we work to address differences of luck when a child is born, but then ignore bad luck during life? I’ve been thinking a lot about luck lately, and have a book coming out later this year on the topic. The more I delve into it, the more I notice that luck shapes a great deal of life’s outcomes.

Interestingly, it doesn’t require a veil of ignorance to find support for egalitarianism. Surveyed about their ideal distribution of wealth in Australia, even the most affluent respondents nominated a distribution that is more egalitarian than the one we have today. Asked how much wealth the top quintile should have, they nominated a figure of 24 percent, considerably below the actual figure of 62 percent.[8] Asked whether ‘differences in income are too large’, three quarters of Australians agree, up from two-thirds in the mid-1990s.[9]

Scholars studying happiness frequently ask respondents to rate their life satisfaction on a 0 to 10 scale, and then aggregate these figures across a society. If we take this perspective – that a one-point movement up the scale is of equal value regardless of the respondent and regardless of the starting point, then we can ask how much money it costs to raise happiness for different people. According to one study, getting a one-point happiness gain for the poorest households cost $6000, compared with $100,000 for the richest households.[10] This suggests that a well-targeted social safety net like Australia’s does a lot to raise average happiness – even if the bucket leaks along the way.

I once asked my 8 year-old Sebastian which of the following he would prefer. If he got two biscuits and his 5 year-old brother Theodore got three. Or if he and Theodore got one biscuit apiece. He replied immediately that he would prefer the latter, ‘Because it would make me sad if Theo had more than me’.

You might chuckle at this, but plenty of economic research finds similar results among adults. A US survey found many people would prefer to be poorer in absolute terms, so long as they were better-off in relative terms.[11] In the ‘ultimatum game’, many players will forego money rather than accept a result that they regard as unfair. In most contexts, the threshold for unfairness in the ultimatum game is when one person gets more than three times as much as the other.[12] Psychology researchers have documented a strong tendency to barrack for the underdog.[13] Across neighbourhoods and between countries, those in unequal places tend to be less happy with their lives. This reflects what economist Lester Thurow once called ‘the income distribution as a pure public good’.[14]

Instrumental concerns about inequality

The other approach to inequality is to focus on how it affects other things that we care about. In The Spirit Level, Richard Wilkinson and Kate Pickett argue that inequality affects a plethora of social variables. If Britain became as equal as Scandinavia, they argue, mental illness would be halved, teen birth rates would fall by two-thirds, homicide rates would fall by three-quarters, and everyone could get seven extra weeks’ holiday a year.[15] It’s a heady concoction, but I can’t quite bring myself to swallow it.

Much as I appreciate their focus on inequality, my own analysis of the data suggests that when you move away from looking at a snapshot in time to analysing changes over time, the effect of inequality on some of these social outcomes is less clear.[16] For example, one of the sharpest falls in crime over recent decades has occurred in the United States, a country where inequality has risen rapidly. We need to be cautious about blaming everything on inequality.

That said, there are some instrumental reasons to worry about rising inequality. As incomes become increasingly concentrated, there is a risk that affluent donors will exert a disproportionate influence over policy outcomes. In the United States, a recent report listed nine billionaire donors currently being courted by the Republican Presidential hopefuls – among them the Koch brothers and Sheldon Adelson.[17]  These donors will likely have a significant impact on the policy positions that these candidates adopt.

As Republican Party boss Mark Hanna put it, ‘There are two things that are important in politics. The first is money, and I can’t remember the second.’ One study found that US Senators’ voting behaviour is very strongly related to the opinions of their richest constituents, but unrelated to the views of their poorest constituents.[18] In an environment where campaign advertising matters, too much inequality may tilt the political playing field away from the median voter and towards the median donor.

Another instrumental reason to be concerned about inequality is because of its impact on social mobility.[19]

To see this, imagine a ladder in which mobility reflects the extent to which a child climbs up or down from their birth rung. In a fully mobile society, the rung you end up on is independent of the place you started. In a static society, people are born and die on the same rung. Most of us viscerally recoil at the thought of such a feudalistic outcome, with the waste of talent that it implies. But yet there is mounting evidence that inequality and immobility go together.

If mobility is the extent to which a person moves up or down the ladder, mobility can be thought of as the gap between the rungs. A society with high inequality – with large gaps between rich and poor – is one in which the ladder is hard to climb. But evidence from across countries and neighbourhoods seems to show that more equal societies are also more mobile societies. Contrary to those who say that we ought to tolerate inequality because anyone can make it, the truth turns out to be that very unequal societies are also those in which fewer people move from rags to riches.

What can be done?

What policy reforms might reduce inequality? In Battlers and Billionaires, I proposed eight ideas:

1. Maintain pro-growth policies
2. Improve our education system
3. Acknowledge the role of family structure in perpetuating disadvantage
4. Recognise the role of unions in reducing inequality
5. Ensure that our welfare spending is targeted to the neediest
6. Maintain a progressive income tax system
7. Make more use of randomised policy trials to evaluate social policies
8. Keep egalitarianism at the heart of our national story

Today, let me add four new ideas to that list that I think are worth considering. In the interests of full disclosure, let me also admit that three of them are drawn from Tony Atkinson’s excellent new book, Inequality: What Can be Done?, which I heartily recommend to you.

9. Put new policy ideas under the equality lens. With inequality at a 75-year high, no government can afford to ignore the implications that new policies have on inequality. At present, new projects and policies might be subject to Regulation Impact Statements, Regional Impact Statements, Environmental Impact Statements, and Health Impact Assessments. Yet there is no similar process to look at distributional effects. Whether such a process should be formal or informal is an open question, but closer scrutiny of inequality ought to be on the agenda.

10. Encourage ethical behaviour by firms and executives. In the wake of the Global Financial Crisis, Harvard’s MBA class of 2009 originated a voluntary pledge for graduates, in which they promised to ‘create value responsibly and ethically’. In the employee-owned company John Lewis, a pay policy does not allow the highest-paid person to get more than 75 times what the lowest-paid person receives. In TSB Bank, the figure is 65. In Traidcraft, it is six. If this sounds radical, recall that Plato thought the society-wide figure should be four.[20] While I do not favour a compulsory cap on top salaries, I do admire firms that are beginning a conversation about pay equity within their ranks.

11. Consider inequality in competition policy. In arguing for competition laws in the United States, Senator John Sherman, one of the originators of competition policy, cited excess inequality as one of his concerns, worrying that ‘inequality of conditions of wealth, and opportunity that has grown within a single generation out of the concentration of capital into vast combinations to control production and trade to break down competition’.[21] And yet our competition law is silent on the issue of equity, with its object merely being ‘to enhance the welfare of Australians through the promotion of competition and fair trading and provision for consumer protection.’[22] A more explicit consideration of inequality in competition policy would not change most outcomes, but might make a difference on the margins, with considerations such as the distribution of outlets in low-income neighbourhoods.

12. Recognise the relationship between economic inequality and gender inequality. According to figures released earlier this year, the gender pay gap is now at a 20-year high. Part of the reason for this is that many of the lowest-paid occupations (eg. child care worker, hairdressers and cleaners) are majority-female; while many of the highest-paid occupations (eg. surgeons, financial dealers and actuaries) are majority-male.[23] As the pay gap between child care workers and surgeons has grown, the gender pay gap has grown too. In this sense, economic inequality is a feminist issue. Related to this is the gender wealth gap. Within the top 1 percent, only one-fifth of all wealth is held by women; a share that drops to one-thirtieth if Gina Rinehart is excluded.[24] Better recognising the way in which economic inequality impedes gender equity in Australia is an important facet of addressing inequality.

Over recent years, inequality has been taken up as a central issue by many senior figures.[25] In Osawatomie, President Barack Obama drew on Theodore Roosevelt’s example to highlight the importance of the issue. In Evangelii Gaudium, Pope Francis denounced inequality and called for action on social disadvantage. Last year, Thomas Piketty’s 700-page book on inequality topped the best-seller lists, sparking a global conversation, while giving heart to economic historians toiling in dusty library basements everywhere.

Between Piketty, the Pope and the President, inequality is fast becoming a central issue of our age. Pen’s Parade reminds us that the disparities between rich and poor are significant. Inequality matters for instrumental reasons, but mostly for intrinsic ones. Put simply: most of us want a more egalitarian Australia than we have today. A richer conversation about inequality is not only in the interests of the disadvantaged, but all Australians who want to maintain a fair society.