Corrado Gini (pictured) was a creative and wide-ranging sociologist, demographer and statisti- cian—but not the sort of human being who is remem- bered fondly. Photo: Prospect composite

Rebottling the Gini: why this headline measure of inequality misses everything that matters

Capitalism is killing less fortunate citizens. We will never understand how if we continue to reduce discussion about the economic gap to a single—and flawed—statistic
January 27, 2020
Corrado Gini (pictured) was a creative and wide-ranging sociologist, demographer and statistician—but not the sort of human being who is remembered fondly. Photo: Prospect composite

Concern about inequality is growing—at least among those who don’t count themselves as economists. It has been cited as a reason for Brexit and the election of Donald Trump, while it has also played a central part in the political campaigns of Bernie Sanders and Elizabeth Warren in the US as well as the Labour Party in the UK. Many economists, especially on the right, argue that inequality is nothing to worry about. Why? Look at the Gini coefficient, they say. While it is true that this headline inequality measure rose quite quickly in Britain when Margaret Thatcher was prime minister, it has actually been pretty flat or even falling for 30 years. The US, they acknowledge, is a good deal more unequal than Britain, but even there the Gini has hardly moved since the financial crisis, and so—these economists say—it can hardly be blamed for the arrival of President Trump.

But these economists are making an error—they are fixating narrowly on measuring only one thing, the spread of incomes, and doing so with a gauge that is flawed. The very fact that many concerned citizens—and no doubt Prospect readers—have heard of the “Gini coefficient” bears testimony to just how much anxiety about inequality there is. In the early days of the 21st century, it was merely one of several measures of statistical dispersion that specialist researchers would use, among other things, to gauge the gaps between incomes. But after the financial crisis this single number suddenly loomed large in blockbuster state-of-society books, such as The Spirit Level, and in seminars in the Obama White House, where it was used to show that America’s problem with social mobility was matched by its problem with inequality.

What really, though, is the Gini coefficient, where does it come from? Corrado Gini was a creative and wide-ranging sociologist, demographer and statistician—but not the sort of human being who is remembered fondly. An eminent Italian economist told one of us that in some circles, no one speaks his name, instead referring to him as “Dr G,” because it is supposedly unlucky to utter his name. Rather like performers who are spooked by Macbeth and will only refer to the “Scottish play,” some Italian statisticians would seem to be blighted by the curse of Gini.

Dr G published his work on inequality in Italian in 1912. His idea was to assess it by looking at the average difference between every pair of people in the population, measured in terms of mean income. If we both have the same, the difference between us is obviously nothing—that is, zero lots of mean income. But if I have everything, and you have nothing, I have twice the mean, and the difference between us is twice mean income, or 2. To fix the index so that it always lies between 0—perfect equality—and 1—the complete inequality of this scenario, we then have to halve this total. That’s essentially it, but the number of two-way comparisons rises rapidly with the population, because each person has to be compared to every other person. With three people, there are three comparison pairs that we tot up differences across, average and then halve; with four people, there are six such pairs. And so on.

You may be beginning to see the allure of the Gini. It offers a single calculation for a single summary statistic which, with its 0 to 1 scale, seems as easy to read as a percentage, and which—to the uninitiated—appears to tell you all you need to know about who is getting what income. It is flexible, too. It can be applied not only to income—about 0.35 for Britain, and about 0.47 for the US—but also to wealth, where the Gini is much larger, because differences in income persist and are cumulated into wealth over the course of life and over time by family dynasties. Indeed, it can also be used to describe the spread of anything else—weight, intelligence, or your number of social media followers—that is unequally distributed, not that you would know that from an “inequality debate” that fixates on the finances alone.

The first time Gini introduced his measure in English was in a comment on a 1920 paper by the British economist Hugh Dalton in the Economic Journal where Dalton had introduced the “principle of transfers,” that social welfare would improve whenever resources were transferred from a richer to a poorer person (provided that the transfer did not switch their positions). Dalton went on to become Chancellor of the Exchequer in 1945, in a government that made great steps to implement progressive transfers by laying the foundations of the welfare state.

Gini also held a position of power, but he was on a different political course. As the founding director in 1926 of ISTAT, the Italian Statistical Agency, he met regularly with Mussolini. He was a eugenicist, and the author of a paper called “The Scientific Basis of Fascism.” He helped rationalise Mussolini’s pro-natalist policy and his Ethiopian colonialist adventure, which he saw as an outlet for the Italian population as well as a civilising mission to a backward people. After the Second World War he supported a bizarre scheme whereby America would take over other rich countries and establish a world government in Washington, DC. We are fortunate to have the Marshall Plan, not the Gini plan.

Putting all this together, it seems unlikely Gini had much interest in reducing inequality—especially not when his haughty personal conduct is factored in. Franco Modigliani, the great Italian-American economist, a Jew who left Italy in 1939, told one of us that when he was at a conference with Gini, the older man had handed the younger his watch saying that it was broken, and would he, Modigliani, please see to its repair.

Measure of the man

So much for the flaws of the man, but are there flaws in his measure? For starters, the thing that is most alluring about the Gini coefficient also turns out to be its greatest shortcoming. By collapsing the whole rainbow of the income distribution into a single statistical point of white light, it necessarily conceals much of great interest. That is of course true of any single summary measure of inequality. But measures like percentile ratios are not only more transparent, but tell us more. They might compare—say—the person who is a tenth of the way along the income spectrum with the person who is, say, a tenth of the way from the top (the 90th percentile). If you are told that the 90/10 ratio rose by so much, this will tell you something about the general spread of incomes in the middle, but plainly much less about what’s going on at the real extremes. Conversely, if you take the route popularised by Thomas Piketty and focus on the shares of income and wealth claimed by the top 1 per cent, then it should be obvious there are further questions to ask about what’s happening in and amongst the other 99 people in every hundred. The best measures are those that match our purpose, or pick up on the places where important changes are happening. We should pick and mix with that in mind.

But focusing on the Gini alone, and thinking of it as the measure of inequality, encourages the illusion that one number can convey an overall picture. It also has specific problems of its own even compared with other aggregate measures. Many people worry about the super-rich, so a useful thought experiment is to see what happens to an index when we take the person at the top and give them even more. For the Gini, the way an individual’s income affects the index depends on their rank in the distribution and not directly on their income. If we take money from the poorest and give it to the richest, the increase in the Gini is the same no matter whether the richest person is a millionaire, a billionaire, or a multibillionaire, and this insensitivity is something to which we might reasonably object. If billionaires are our problem, we need someone other than Dr G.

How far we should care about billionaires is a live political question, as the recent UK general election demonstrated, when Labour’s John McDonnell got into a row with one of their number, Phones 4u founder John Caudwell. Even if the Caudwell side of the argument came out on top in 2019, the underlying dispute runs deep, and will not be banished just because it doesn’t register on a particular statistic. Hugh Dalton, like his Labour Party successors today, would have seen billionaires as a prime source of transfers. They have so much wealth that losing some would hurt them very little while the proceeds would benefit poorer people much more. This is the principle of transfers in action. Many economists today would endorse this position, which modern philosophers have come to refer to as “prioritarianism.”

Some philosophers, and a few economists, go further and are out-and-out egalitarians. They like equality for its own sake, and are willing to destroy income and wealth at the top in the interests of a more equal distribution. Emanuel Saez and Gabriel Zucman in their recent book, The Triumph of Injustice, argue for destruction of pre-tax income at the top. Our own take, in line with the majority of economists, is that a super-rich person does not harm any one of us simply by being richer than us, though we certainly have reason to complain if the rich person uses his wealth to harm us, for example by buying politicians to act against our interest. Inequality is not inherently bad, though it can be instrumentally bad, a distinction on which an international league table of Gini coefficients sheds no light.

There are also arguments in favour of billionaires which such a league table would miss. According to Forbes, the richest men—and they are all men—in America in 2019 were (with their fortunes in brackets) Jeff Bezos ($114bn; post-divorce, his ex, MacKenzie Bezos is just outside the top 10), Bill Gates ($106bn), Warren Buffett ($81bn), Mark Zuckerberg ($70bn), Larry Ellison ($65bn), Larry Page ($56bn), Sergey Brin ($56bn), Michael Bloomberg ($53bn), Steve Ballmer ($52bn) and Jim Walton ($52bn). All of them except Buffet built new products or new ways of doing things that improved lives. They got rich by benefiting, not only themselves, but others too. If the chance to get mega-rich was their spur, then taxing them heavily might discourage others from emulating them. The US has always been a great attraction for talented, ambitious immigrants, and some of the lure might be the chance to be a billionaire.

Much more generally, economic development is almost always uneven, with some benefiting before others, so that inequality can sometimes be a sign of progress. This has often also been true in health where new treatments or new medical knowledge—like the dangers of cigarettes—are first made available to or adopted by the better off and the better educated. The emergence of a few non-kleptocratic billionaires in recently poor places, like India, China, Turkey, and some African countries, can also be seen as a positive indicator that the benefits of capitalism, long closed to most of the world, are spreading. In sum—for good or ill—billionaires matter. But the Gini coefficient passes over them in near silence.

The wrong riches

While the question of billionaires is divisive, there are some kinds of unfairness that almost everyone dislikes. Stealing from the rich to give to the poor—Robin Hood theft—can be grand larceny or philanthropy, depending on your values. But no one defends Sheriff of Nottingham redistribution, which reverses the Dalton principle, by stealing from the poor to give to the rich.

“Tech companies look more and more like takers than makers”

How riches are accrued always matters a great deal for fairness, but this crucial aspect of political economy is something that no all-purpose inequality index tells us anything about. Keeping track of who is making and who is taking is tough because makers can turn into takers in the twinkling of an eye. The same tech companies who have brought consumers and citizens so much innovation have lost their lustre, and look more and more like takers than makers. To the extent that relatively more of their profits come from their lobbying of Congress, from their excluding competitors, or from their possession of our information and less from continuing innovation, then those profits will not need to grow—or move the Gini coefficient—to run into growing resentment.

Other American industries are already deep into this territory. The American healthcare system is the most expensive in the world, and delivers some of the rich world’s worst health outcomes. It is also a gigantic engine of Sheriff of Nottingham redistribution, taking money from everyone—poor, middle, and well-to-do—and distributing it upwards to well-heeled doctors and providers in hospitals, pharmaceutical and device manufacturers. A family health policy cost $20,000 on average in 2018, and because so much health insurance is paid as part of employment, this high cost reduces wages and replaces good positions in large firms with the sort of outsourced jobs in service supply firms that offer scant security and few prospects. The cost of healthcare is essentially a poll tax on all Americans.

The iniquity of running healthcare as a profit-making enterprise is perfectly illustrated by a recent plague of “surprise” medical bills. Emergency room doctors and ambulance services in many places have been taken over by private equity firms and taken out of insurance networks. Patients are billed by doctors that they have never met, or ambulance crews who signed them up during emergencies, perhaps even when they were unconscious. When patients cannot choose, they are perfect targets for predatory pricing. Almost everyone, including politicians of both stripes, agree this should be stopped, and it almost was in the budget bill of December 2019. According to the Washington Post, last-minute pressure from powerful New York hospitals was enough to persuade Senator Schumer, the Democratic Minority leader in the Senate, to drop the provision.

The argument about exactly how much inequality—how high a Gini—is needed before injustice sets in will always rage. But there can be no such argument about these disgraceful features of American healthcare. Highway robbery of its citizens—literally so after a road accident—is being protected by the elected politicians who ought to be policing it. The proceeds go, not to possibly-poor highwaymen, but to wealthy investors. The rotten thing in contemporary capitalism is not principally the concentrations of wealth it allows, but the fact that it allows them to develop in such outrageous ways. It breeds rage, cynicism and even—as our new book establishes—loss of life.

A deadlier inequality

Deaths of Despair and the Future of Capitalism describes an epidemic of American mortality over the last two decades, driven by mortality from drug overdoses, suicide and alcoholic liver disease. This epidemic is not caused by inequality, but it is intimately tied to it.

Most obviously, these extra deaths—and the rising physical pain, mental distress and social isolation that have come with them—are not equally borne, but are almost exclusively confined to white non-Hispanics who do not have a college degree, opening up a great gulf in health and life expectancy between the more and less educated. The third of the population with a bachelor’s degree or beyond, the elite that have benefited from economic growth and technical change over the last half century, are almost exempt. So too have been African Americans, at least until fentanyl entered the illegal drug supply after 2013. (Incidentally, African Americans suffered their own, not dissimilar, epidemic 30 years earlier. What happened to the black communities then is happening to less-educated white communities today.) The consequence of the recent trends is that while black Americans still die earlier on average than white Americans, as they always have, over the last generation the gap has closed. Gaps matter, and so this is important. But it is cold comfort to those communities of less-educated whites where the trend in life expectancy has been in the wrong direction.

How and why did this happen? Less-educated white non-Hispanic men have seen their real earnings fall for more than half a century as the number of good jobs for them has declined. For some of them, rising earnings from the women in their households will have compensated for family income, though not necessarily in terms of their self-esteem and mental health. What may hurt more is not inequality per se, but the dashing of the particular expectations that individuals were raised with—that they would do better than their own parents, just as their parents had done better than their parents. Once again this is something that neither the Gini, nor any other mathematical measure of inequality, can tell us about.

“Many less-educated men—and more recently women—have dropped out of the labour force altogether”

Many less-educated men—and more recently women—have dropped out of the labour force altogether; many others have lost good jobs with large firms, and are now working in outsourced jobs that supply labour to the large firms that used to employ them. Because “good employers” provide medical cover for their workforce, this long-term disintegration of the working-class labour market is in no small measure attributable to the “absurd and oppressive” (in Adam Smith’s phrase) cost of healthcare that we have described. Its financial consequences pour over and into problems of other sorts, because the new jobs often lack healthcare and pensions and the security that they bring, as well as any opportunities for promotion or the meaningfulness that comes for working in a flourishing enterprise.

Worse, people with poorer prospects are less marriageable, and marriage has fallen dramatically among less-educated Americans. So has fertility, though many less-educated couples engage in unstable serial partnerships, often resulting in children, so that there are men in midlife who lack the comfort and stability of a lifetime relationship, and who do not know their children, who are living with other men.

American society, then, has translated unequal educational attainments into gross and deadly disparities in day-to-day family life in ways that cannot be either captured nor foretold by looking at a single aggregate inequality measure, such as the Gini. Meanwhile, the healthcare industry, which is much more successful at generating wealth for providers than health for its patients, is profiting handsomely from the misery. Again, the problem is not just the skewed distribution of the riches, but the way that these riches are being amassed.

America’s communities of despair are a fertile ground for an iatrogenic epidemic of opioids, supplied by pharmaceutical companies, approved by federal regulators, and dispensed by initially careless doctors who were persuaded that opioids—essentially legalised heroin—were a suitable remedy for chronic physical and social pain. Pharma companies and drug distributors got fabulously rich through the deadly addictions that followed. The Sackler family, whose respectability was attested by two honorary knighthoods for philanthropy, own the privately held Purdue Pharma which manufactures OxyContin, which profited by more than $11bn. Meanwhile, Tasmanian Alkaloids, a subsidiary of Johnson & Johnson, the company that makes Band-Aids, Baby Powder, and Tylenol, cooked up a genetically altered opium poppy grown in Tasmania that provided the raw material for the epidemic. Just as the East India Company got rich by exporting opium to China in the early 1800s, the US pharmaceutical industry got rich through the addictions and deaths of less-educated Americans. Now, as then, the companies used politicians to protect their profits.

Despair fostered not only the drug deaths, but also the rapid increases in suicide and in deaths from alcoholic liver disease whose numbers, taken together, are even larger. The behaviour of pharma and their regulators threw fuel on the flames, but the underlying despair grew out of a half-century of economic progress for the well-educated that did nothing for those without a college degree. It isn’t income inequality that is killing so many Americans, but a much broader, long and grinding decline affecting less-educated Americans and their communities.

Better-educated Americans have prospered from the same globalisation and automation that endangers the working class. To add insult to injury, an absurdly costly healthcare system redistributes resources upwards, aided by an iniquitous political economy that supports simultaneous and parallel increases in inequality and mortality. You could crunch the Gini coefficient to as many decimal places as you like, and you’d learn next to nothing about what’s really going on here.

Transatlantic rage

Underneath the recent headline of faltering progress in life expectancy in the UK, we can see that deaths of despair are rising in England too. We do not know the educational attainment of those who died, but it is clear these deaths occur predominately outside of London and in the less successful parts of the country. Thankfully, much better control of prescription opioids has—at least so far—limited the total numbers of deaths to a fraction of those in America, though illegal drugs are a problem for all countries, and the rocketing drug deaths in Scotland are deeply worrying. Britain is fortunate, too, that in the National Health Service, it has a much less costly healthcare system, and not one that undermines jobs by tying healthcare to employment.

“Until the crash, it was possible to believe that the salaries of the bankers were extreme but in the public interest”

But there is another problem that British and American capitalism do have in common—less deadly than despair, but every bit as poisonous for our politics. Namely, the legacy of the financial crisis. The crisis did not contribute directly to deaths of despair in the US, which climbed inexorably before it, through it, and after it. And if anything, it slowed or brought an end to rising income and wealth inequality. Even so, it is hugely important for understanding the current anger about inequality.

Until the crash, it was possible to believe that the salaries of the bankers, the pharmaceutical executives, the hedge fund managers were indeed what we were told—extreme, maybe, but in the public interest. Those people, it was said, were creating jobs and wellbeing for all of us. The financial crisis, with its unemployment and loss of livelihoods for the many, without punishment or accountability for those few who caused it, ripped away the pretence: Gotterdämmerung for trickle down. And when people saw it in its true light, they were newly angry about the old inequality. While the Gini coefficient was flat, and showing nothing, rage about unfairly shared spoils began to bubble over. Yet more evidence, if it were needed, that a century after Mussolini’s statistician tried to boil inequality down to a single figure, it’s time to declare that Dr G’s number is up.

[su_box title="What the Gini misses, by Tom Clark"]In recent years, the argument about inequality has come to centre on a single statistic—the Gini coefficient. It seems a strange thing to argue about when, as the chart shows, it has been as flat as a pancake in Britain since the financial crisis.

Nothing to see here: the Gini for disposable incomes since 2008 Nothing to see here: the Gini for disposable incomes since 2008

Nothing to see here: the Gini for disposable incomes since 2008

But just because one gauge of one dimension of inequality hasn’t risen, it doesn’t follow that it’s wrong to worry about the economic gap. All sorts of other data reveal important respects in which the rich truly have got richer, and the poor poorer, while at the same time, the great gulf in health between prosperous and penurious communities has been widening too.

article body image

Runaway riches

The cash increase in the total “single figure” remuneration (including bonuses etc) for the median FTSE 100 CEO between 2009 and 2017, which took the annual total to £3.9m.

article body image

Harder times

The increase in the number of Brits living below the main official poverty line (drawn at 60 per cent of the median total post-tax income) over the last four years of data, to take the total to 11.1 million.

article body image

Unequal lives

The widening of the gap in female life expectancy between the most- and the least-deprived English communities, in the last three years of data, which took the total longevity gap for women to 7.4 years. For men, the total gap is even bigger, at 9.4 years.[/su_box]