Impunity – an atlas of unfairness

All actions have consequences. It’s a law of physics – Newton’s third of motion – and it also ought to be a rule in human societies. Generally it is, but as unfairness wields its influence, the rich and powerful are able to avoid many of the consequences of their activities.

That’s the basis for a new atlas of impunity, produced by the Chicago Council on Global Affairs and the political risk consultant Eurasia Group. As they put it:

“Impunity is the exercise of power without accountability, which becomes, in its starkest form, the commission of crimes without punishment.”

The report goes on to suggest that: “Impunity thrives when the imbalance of power is so great that the powerful think they do not have to follow the rules.” In other words, impunity is a clear form of unfairness. And unfortunately, “This feeling is one we think is prevalent, and which we fear is on the rise.”

There are detailed conclusions but the headlines are summarised in this global map:

The published data covers 197 countries in total, of which the authors believe they have enough data coverage to rank 163. The overall analysis represents a new tool for businesses and investors in thinking about the world and the stability and reliability of its regimes. The analysis of how such tools might be deployed by both investors and companies in my recent paper for the Bingham Centre for the Rule of Law applies.

Indeed, there’s significant alignment between the insights from the Atlas and existing global assessments of the Rule of Law. Overall, there are 8 indicators specifically from a Rule of Law analysis (the World Justice Project’s Rule of Law Index) and many others will be recognisable as measures used in such analyses. As with other measures of the Rule of Law, weaker performance reflects a poorer environment for business activity and so a brake on overall economic performance.

The 8 explicitly Rule of Law indicators are out of 67 in total, and are spread across 3 of the 5 components of the overall index. These components are:

Unaccountable governance
Abuse of human rights
Economic exploitation
Conflict and violence
Environmental degradation

It is here that the Atlas of Impunity differs from previous measures of Rule of Law and similar assessments: within the environmental and conflict components, the Atlas includes elements considering a country’s impact on the wider world. Both components are predominantly about issues within countries, such as air quality and waste management, and community safety and homicides respectively. But they also include indicators including (again respectively) the overall ecological footprint and involvement in the arms trade. As the report states, “the impunity framework helps us understand the connection between what happens at home and what happens abroad. It speaks to interdependence in a way that national frameworks cannot.”

This seems fair and an important advance on what came before. It brings in a sense of a nation state’s impunity for the damage that it causes the rest of the world, a sense of the fairness of its influence on the wider world. However, it is notable that the correlation between these two elements and the overall score is much lower (0.80 for conflict and violence and 0.69 for environmental degradation) than the remarkable levels of correlation between the other three elements and the overall score (between 0.93 and 0.96). So though Norway has low scores on the environment because of its export of fossil fuels and its rich-world environmental footprint, pulling its scoring notably below its neighbours Finland, Denmark and Sweden, it still ranks fourth overall behind these three. Canada’s position is also notably weakened for reasons similar to Norway, but still ranks highly.

The list of the weakest-performing countries is perhaps even less surprising: Afghanistan, Syria, Yemen, Myanmar and the Central African Republic, all of them scarred in different ways by their recent pasts. The authors are keen to point out that while results are framed by the past, history isn’t destiny. They note Senegal and Ghana as among those outperforming their difficult heritages.

As with other such indices and rankings, there is much that both companies and investors (in their sovereign debt investment and stewardship approaches, but also in corporate stewardship activities) can do to reflect on the lessons from them, and work to reinforce better and more robust societies in the countries in which they invest. That will bolster economic prosperity and so business and investment performance, as well as helping to build a fairer world.

Board actions to protect value and boost fairness through Rule of Law

Atlas of Impunity 2023, Chicago Council on Global Affairs and Eurasia Group, February 2023

The Rule of Law and investor approaches to ESG: Discussion paper, Bingham Centre for the Rule of Law, September 2022

World Justice Project’s Rule of Law Index

Impartial – or fair?

Our modern society convulses on occasions about issues that burst into flame from seemingly remarkably small sparks. It’s somehow a feature of the social media age, and perhaps of the nature of politics that social media seems to encourage.

In the last weeks in the UK, one convulsion has been in relation to a brief tweet by a former footballer criticising the language surrounding one aspect of government policy. It’s more complicated than that, of course: the former footballer is Gary Lineker, who leads football coverage at national broadcaster the BBC (and is its highest paid star), and the policy is one of the new Prime Minister’s 5 key promises to the nation, to stop the arrival of a few thousand individuals by boats across the Channel from France.

Lineker, who famously never received a yellow or red card as a player, was sent from the pitch by the BBC, only subsequently to be reinstated after a wider rebellion by pundits and commentators neutered its sports coverage. At its heart the BBC’s convulsion was in relation to its commitment to ‘impartiality’ and the extent to which that commitment should limit comments by contractors such as Lineker, particularly those operating in non-news areas of its coverage. No one disputes that employed news staff must avoid sounding personally political.

I don’t intend myself to discuss government policy nor the words Lineker chose to use, nor other aspects of alleged politicisation at the BBC. A lot has already been said about the fairness or otherwise of Lineker’s treatment and the treatment of other BBC contractors. Rather than turning that over again, I want to explore what the BBC and others think is meant by impartiality and whether it’s the right concept to apply at all. I am grateful in this to the inspiration of a short post by legal blogger David Allen Green, which considered the concept of impartiality as it is applied in the courts.

A key section of Green’s commentary is the following:

“the judge is required…to decide the dispute impartially and having given each side a fair hearing./What the judge will not do – even though they are duty-bound to be impartial – is to treat both sides as having equal weight and not make any material decisions at all.”

He rails at what he and others sometimes call the ‘both-side-ism’ that the BBC can appear prey to: the tendency occasionally simplified as reporting two points of view, one that it is currently raining and the other that it is not, rather than looking out of the window to verify the situation one way or the other. Of course, most issues are more nuanced than the presence of precipitation (even in the context of the UK’s propensity to wide varieties of rainfall). In a more substantive area, many complain that the BBC used to have both climate scientists and climate change deniers weigh in on the issue, as if the balance of opinion was somewhere halfway between these, when the unanimity of scientists about the climate crisis is almost total. Happily the broadcaster seems to have gained greater confidence in the certainty of the science of late.

The oddest thing when exploring impartiality at the BBC is that though this is the term that has been used as a one-word drumbeat throughout the recent debates – the Director General’s comments were littered with that single word – it is not what the BBC is committed to, nor what the regulator, Ofcom, insists on. The broadcaster is called on not to apply ‘impartiality’ but ‘due impartiality’. This is the term used in Ofcom’s Broadcasting Code. It’s also the relevant term in the BBC’s foundational document, its Royal Charter, and in the Framework Agreements between the organisation and the government that sit beneath the Royal Charter. The BBC is not expected to be impartial; it is required to be duly impartial. What’s more, that requirement applies largely only to its news reporting: the word ‘impartial’ appears only to be applied within the foundational Royal Charter with regard to the BBC’s “news, current affairs and factual programming”.

It’s been reinforced more recently in the latest of the so-called Framework Agreements that are the vehicle for the government to assert its expectations of the BBC. The 2022 Framework Agreement inserts clear obligations for the BBC board, under the unequivocal banner “The board must:”. Bullet (b) says that the board must “ensure in particular that any such guidelines set appropriate standards to secure the fairness, due impartiality, due accuracy and editorial integrity of the content of the UK Public Services”.

In its Broadcasting Code, Ofcom clearly explains the importance of the word ‘due’:

““Due” is an important qualification to the concept of impartiality. Impartiality itself means not favouring one side over another. “Due” means adequate or appropriate to the subject and nature of the programme. So “due impartiality” does not mean an equal division of time has to be given to every view, or that every argument and every facet of every argument has to be represented.”

So the BBC is explicitly not required to be impartial, and should not be seeking to apply that standard, it must be duly impartial. As Ofcom explains, that means it must apply appropriate judgement and be impartial with a wise sense of the context and of reality. In short, it must be fair – it’s notable that many dictionaries offer fairness as an alternate term for impartiality. But surely fairness is an even better synonym for due impartiality: both terms require sense and judgement, and would expressly lean against lazy both-side-ism.

The banner for Section 4 of the BBC’s editorial guidelines

It’s sad then that the BBC’s own Editorial Guidelines are casual in how they consider the term impartial. Section 4 of the Guidelines is entitled Impartiality, not Due Impartiality. Further, while the introduction of the section does talk about due impartiality, the explanation seems less precise than Ofcom’s and in fact seems to invite the introduction of all points of view, just not necessarily in a single programme but instead potentially over a whole range of programming. It feels like an invitation to both-side-ism rather than a clear expectation that journalistic judgement must be applied and fairness should prevail. Later in the section, both ‘impartiality’ and ‘due impartiality’ are used, without any apparent consciousness that they are different things and that the BBC is not required to be ‘impartial’ without the context and sense that comes with the ‘due’.

Perhaps unusually for this blog I’m not seeking to argue that the term fairness should be substituted for due impartiality – that term seems well-embedded and well-understood at least by the BBC’s regulator if not so clearly by the broadcaster itself. But I do think that the judgement and sense that is freighted in the term fairness needs to be learned from. Due impartiality is not impartiality – it is a much better and more demanding standard. It is not fair to its great heritage that the BBC is lazily drifting from living up to the standard of due impartiality and apparently trying to substitute the weaker term of impartiality.

A fully equal society: The Dispossessed

When recently reading The Dispossessed, Ursula Le Guin’s 1974 multi-award-winning sci-fi novel, I kept on coming back to thoughts of the totally equal society theorised by Martin Moryson in his paper proposing that there is an optimum level of inequality for growth (see Optimum Inequality?).

For The Dispossessed features a planet that comes close to delivering on that total equality: Anarres is a land where personal possessions are frowned on, food is shared and eaten communally, and all are expected to give some of their time to menial labour for the benefit of the community. Possession is so frowned on that there are no possessive pronouns: most disconcertingly, there is no way to say “my mother”, she is just referred to as “the mother”, with all the limits to emotional depth of feeling that this implies. Indeed, children are more a community asset than members of a nuclear family. The worst insult on Anarres is ‘propertarian’, a term frequently deployed to shame selfish and anti-communitarian behaviour.

I don’t think it is a failure of Le Guin’s imagination – she had one of the richest and most generous imaginations – that she sees this spirit possible only in a world of dearth and hunger. Anarres has such sparse resources that there is simply no scope for any individual to gather wealth or a disproportionate part of the resources of the planet. It does feature one species of tree (of course, because Le Guin was obsessed with trees) but almost nothing else that is green. It is a dusty, dry planet, panicked by drought and famine a few years back. Its vaunted community spirit didn’t stop some communities commandeering food as it was being transported through for the benefit of others during those hardships. But it is only through that community spirit, sacrifices of individuals to the wider good, and acceptance of hard labour for little reward, that anyone survives on the planet at all. There is no surplus to be possessed by anyone, and not much that could ever amount to personal property, other than the most immediately portable items.

Anarres has a twin planet, a green and abundant one, called Urras. One of the countries on Urras is where the true propertarians live. This is not far from Moryson’s maximally unequal society: some enjoying lives of extraordinarily wasteful consumption and luxury while an underclass suffers and scrapes a bare living. Violence, implicit and sometimes horrifically unleashed, keeps this underclass in check and under constraint. With its controlled media and near constant war with other countries on the planet, there are strong echoes of Orwell’s Nineteen Eighty-Four.

Urras is clearly an unfair place. But as I suggested in Optimum Inequality?, neither is the maximally equal society of Anarres fair. The very human trait of personal ambition is constrained, crushed out by near-brainwashing during schooling and in adulthood challenged by social norms and directly in public meetings. Those who refuse to comply are squeezed to the fringes of society, and sometimes deemed mad and committed to institutions. People are not allowed to flourish to their greatest, which seems distinctly unfair. It’s why I don’t think it’s a limitation of Le Guin’s imagination that makes Anarres such a desolate place: only where there is no chance of surplus, she implicitly says, will human beings accept such constraints on their personal lives and scope for personal advancement, including hopes for a better life for ‘their’ children.

If we want a fair world – and one of the central precepts of this blog is that humans do, viscerally and by our very nature, want a fair world – we need to navigate the realities of abundance but find better ways of sharing that abundance than we currently do. It would not be fair, it would not be human, to have as equal a society as that of Anarres. But we need to move nearer towards it than we currently are, which is closer to the unfair inequalities of Urras – both to deliver greater economic growth, as Moryson argued, and to fulfil our collective and visceral need for a fairer world.

By the way, if you haven’t read any Le Guin, I wouldn’t start with The Dispossessed. I’d start almost anywhere else, but certainly with the Earthsea books and with her short stories, collected as The Wind’s Twelve Quarters and The Compass Rose. Among her sci-fi novels, the extraordinary The Lathe of Heaven is my personal favourite. In all of her best fantasy and sci-fi creations, of distant and different worlds, she brings us back to a closer understanding of what it is to be human, with deft and quiet moments of beauty and gentle kindness. And there are always trees.

Optimum inequality?

It’s undeniable in the modern world that some countries are too unequal. Take the seeming inability of South Africa to escape its past – it remains an outlier in terms of its Gini coefficient (economists’ usual measure of unfairness) as it hasn’t yet managed to share the benefits of social and political change with the bulk of the population. But are some other countries too equal, to the detriment of their economic growth? That’s a recent argument from Martin Moryson, chief economist Europe at DWS Research Institute, a thoughtful organisation attached to Germany’s largest fund manager.

Moryson posits that there must be a growth-optimising level of inequality from basic principles: in a totally equal society, no one would have any incentive to work as all would have the same and earn the same whatever they did; similarly, in a wholly unequal one, with all value accruing to a single individual, others would have little reason to do anything other than what they were forced to do. Both theoretical wholly equal and wholly unequal societies would thus struggle to grow at all; somewhere between these two extremes must lie an optimum level of inequality that would lead to the greatest possible growth for that society. As an aside, note that the perfectly equal society Moryson theorises from is just as wholly unfair as the perfectly unequal one – as ever, this blog argues that fairness is a much more useful, and human, lens for looking at the world than mere inequality.

Some will recognise in this a shadow of the argument that drives the more than somewhat discredited Laffer curve, which has been used as an argument that lowering taxation will lead to an increase in tax receipts. That theory has not tended to be reflected in the reality of countries that have attempted so-called trickle-down reforms. Let’s not allow that to be counted against Moryson’s work.

He bolsters his theory with a regression analysis based on data from the last 40 years. This suggests that there is indeed a level of inequality that is optimum for economic growth. This is represented by a Gini coefficient of 30 or just above (based on household income after taxes and transfers):

Wisely, Moryson in discussion makes clear that he does not hold to this number with absolute precision and that the cluster of countries with post-redistribution Gini coefficients of just below to a little above 30, are all likely to be in a sweet-spot for their economic growth. That sweet-spot would in the language of this blog be the point of fairness, where in contrast to the perfectly equal and unequal societies all citizens feel that they have sufficient participation in the economic life of their nation that they are able and willing to contribute.

Outside that fair sweet-spot, it is those countries placed at the greater extremes that perhaps have something to consider and to learn from this analysis. Those with such opportunities may perhaps be best revealed by this chart from a different recent paper on inequality (whose focus is Germany, hence the red shading in this chart, which is not relevant for our purposes):

So, by Moryson’s argument, Slovenia (see also Fairness is a Choice II) is seeking excess equality through its taxes and transfers, proportionally one of the largest in the world. It could enhance its annual growth rate by around 0.1% if it reduced its redistributive policies. Actually, Moryson is wiser than this. He rightly says that Slovenia and its small group of fellow countries whose redistributions take their Gini coefficients down to the lowest levels globally are making a political choice – implicitly rather than explicitly – to favour equality over economic growth. The evidence from Slovenia is certainly suggestive that its population favours that political choice.

There are many more countries that have post-redistribution Gini coefficients well above Moryson’s posited optimum level – including the UK and US, together with South Africa and several other emerging economies. In a sense this must be a political choice too – though giving up economic growth to be more unequal feels more like a failure to make political choices. Redistribution may be part of the steps these countries might take to enhance their growth prospects, Moryson argues. But he also notes – especially for the emerging economies – the vital importance of investment in human capital to build growth capacity, and the central role of social mobility to unlock greater equality (and fairness) in societies.

In a final element of his paper, Moryson constructs an index of inequality, assessing the largest 20 economies across 5 different measures. Even though India benefits from being a democracy (one of the 5 elements of the index, because Moryson argues that it makes any inevitable inequality more tolerable), it is by far the most unequal in the ranking, well beyond Indonesia and China, the next worst performers.

At the upper end are the Netherlands, closely followed by Switzerland, Germany and Canada. Moryson notes that all his 5 measures are correlated but by far the dominant factor, and the one with the highest correlations across the piece, is social mobility.

Countries with sclerotic levels of social mobility will struggle to generate the equality, and fairness, that will help generate greater growth. They are certainly not optimising either inequality or fairness.

ESG Special – Inequality: Inequality – An Investors’ Perspective, Martin Moryson, DWS Research Institute, November 2022

Inequality revisited: An international comparison with a special focus on the case of Germany, Maximilian Stockhausen, Judith Niehues, Institut der deutschen Wirtschaft, May 2021

Fairness is a choice II

Some British people have been shocked recently to learn that the nation’s economic malaise is leading its citizens to be poorer than some from the former eastern bloc. “On present trends, the average Slovenian household will be better off than its British counterpart by 2024” reported a Financial Times story late last year.

Slovenia has a further major advantage: it is a much fairer society. According to World Bank data, only Slovakia has a lower Gini coefficient among all 60 countries for which 2019 data is available. And like Sweden, the country provides evidence that fairness is a choice.

Slovenia was behind the iron curtain during the Cold War, and experienced the Yugoslav version of socialism. This saw different economic phases and differing approaches over time to fairness and equality, but as the country gained its independence and was freed from the shackles of the communist era it took a different route from other countries in the eastern bloc. Rather than adopt the wholesale reforms and abrupt shock treatment urged by the Washington consensus, Slovenia took a more gentle route into capitalism. As a more advanced economy already, it was able to argue that less of a shock was necessary.

That less abrupt approach allowed the country to maintain more fairness than its fellow eastern bloc nations – many of which saw an emergence of an oligarch class and of rampant inequalities. It was a painful period for Slovenia too, but – through a range of different policy choices and the influence of powerful unions – a less unfair one. And as a recent study shows, including a historic view of the country’s Gini coefficient (economists’ favoured measure of inequality), when even Slovenia’s more gentle route towards capitalism led to greater unfairness, the country chose political intervention to reassert a more fair society:

A new progressive income tax in 1994 curtailed the stark growth in inequality unleashed in the post-Yugoslav era. And on those occasions since then that unfairness has begun to grow in Slovenia, there has again been political action to lean against it and maintain levels of fairness that are almost unique in the world – and are enjoyed alongside growing wealth for the nation as a whole.

Over the most recent period for which data is available, the study shows that Slovenia enjoyed some of the fairest growth anywhere in the world. In the period 2008-2018, when many countries saw the income growth enjoyed by the richest outstripping that of the poorest by a margin, the country experienced something very different. Over that decade, the highest-earning tenth in the country enjoyed growth just over 20%, and that seen by all those in the top 60% by income was between 20% and 22%. The poorest tenth benefited from 23% growth in earnings; only the second poorest tenth benefited slightly less, with 18% growth. Fairly shared growth is possible. Fairness is a choice.

As with my conclusion on Swedish fairness, if the Slovenes actively chose fairness and have delivered it, alongside strong economic growth, so can others. We just need to choose fairness.

Be less camel

The desert can be greener than you expect. Certainly, it was greener than I expected.

That was what I saw on a recent visit to the Dubai Desert Conservation Reserve. Our guide explained why. The Reserve doesn’t only restrict access for humans. The vast surrounding fence also keeps the camels out.

Camels have been taught to be greedy over generations of domestication by humans (a heritage estimated at over 3000 years). That means rather than being fair, they simply take all the food available to them, and often eat the whole of growing plants, roots and all. In contrast, wild animals – in the Reserve’s case, Arabian oryx (formerly extinct in the wild but now deemed only vulnerable) and gazelles – tend to be fair to others and to the future by only browsing from the plant life, allowing future regrowth.

This contrast is in spite the fact that the camel, famously, can store large amounts of food for its future. The influence of humans on the animals’ domesticated brain appears to have been pernicious.

Without the unfair feeding activities of camels, the Reserve is therefore green as well as sandy yellows and reds, as the photo shows.

And the lesson for all of us seeking to be fair to others and to the future: be less camel.

Unfairness in carbon emissions

Like so many things in our world, there is an unfair distribution of carbon emissions. That’s clearly true between countries but it’s true, starkly, within countries too. In emissions terms, not all people are created equal. As we come to think about paying for climate change mitigation and financing a just transition to the decarbonised world that we need, those differentials will increasingly matter. The rich will need to pay more; fortunately, they can afford to.

While some measures of per capita carbon intensity focus on fossil fuel producing nations, for fairness we should really be thinking about measures of intensity based on consumption. As a recent academic study trying to understand the differing carbon footprints of consumers puts it: “It is widely accepted as a basic principle of fairness that those benefiting from an activity, like the GHG emission that drive climate change, should bear some responsibility in mitigating the damage caused by those activities.” Those who benefit from the use of goods should shoulder the burden of the carbon emissions associated with their production.

It’s for this reason that arguments that population growth is the most significant driver of increases in carbon emissions are largely wrong. While population growth doesn’t make the challenge of attaining the absolute drops in emissions that we need, it isn’t the fundamental driver, for the simple fact that the countries with the greatest growth in population are among those whose people consume the least and so have lifestyles responsible for less emissions. That’s particularly true following the recent news that China’s population has already started to fall. China is the most carbon polluting nation in the world, though that is mostly due to its (albeit diminishing) role as factory for the world. It is lower in the rankings on a consumer measure of carbon intensity, but actually on a per capita basis it is above the average across the world (7 tonnes a head against 4.7 tonnes, according to Our World in Data).

The range in this consumer intensity measure across the world is remarkable, from 25 tonnes in Qatar and over 15 in the US, to just 1% of that US number – 0.13, 0.16, 0.17 and 0.17 tonnes – per head of population in Malawi, Rwanda, Uganda and Ethiopia respectively. Each individual in India, perhaps already the world’s most populous nation, is responsible for just a tenth of the emissions attributed to a citizen of the US; two other countries with large and growing populations, Indonesia and Nigeria, are also well below the global average, at 2.3 tonnes and 0.6 tonnes respectively.

These ranges are stark, but the ranges within countries are if anything more remarkable. That’s where the recent academic study comes in. Lifestyles and levels of consumption drive striking differences. As shown in the chart above, the academics estimate the carbon footprint of different income groups across the US, and reveal a range in emissions from 17 tonnes for the poorest households to 950 tonnes for the wealthiest 0.1% (note that these numbers are not directly comparable with those above as they are household rather than individual figures). That doesn’t tell the whole story, though: there are a group they dub ‘super emitters’, just 1.5% of the wealthiest 0.1% – around 1900 households – whose emissions are in excess of 3000 tonnes. The recent pictures of the ranks of private jets at Davos remind us that some people live in a different world – or live taking an unfair portion of the resources of our shared world.

Referencing a 2021 paper, which suggests that each US billionaire has a carbon footprint in excess of 8000 tonnes, the academics argue that their estimates for super emitters “are reasonable and possibly conservative”. They certainly seem more reliable numbers than the remarkable 3 million tonnes of carbon which Oxfam America attributes to billionaires in its recent Survival of the Richest report on inequality (the charity appears to attribute significant footprints based on investment holdings of fossil fuel businesses, rather than base its numbers on household consumption).

Andrew Fanning of the Doughnut Economics Action Lab and Leeds University has produced this striking visualisation of the academic study’s results:

Unfortunately, this unfairness in emissions is getting worse. The study notes that in 2019 the average US household was responsible for the emission of 41.7 tonnes of CO2, a reduction of 16% from the 1996 figure. All income groups show a consistent reduction in their carbon footprints – with the exception of the richest. They suggest that the top 1% were responsible for an emissions increase of 23% over that same period – with the top 0.1% behind the bulk of that, because their emissions went up by fully 50%.

The authors make a link to the ‘loss and damage’ agreement that was the sole real progress made at the COP 27 climate conference. Under this, wealthy nations undertook to pay funds to support poorer countries already suffering significant physical impacts of climate change. The richest acknowledged that they are responsible for the bulk of the change in climate, and need to pay to reflect this; the richest within countries, the academics suggest, should similarly step up. They can afford to, because while there is remarkable inequality in emissions, this inequality is less than that for income: “The Gini coefficient for the emissions distribution is 0.35. For comparison the US income distribution Gini coefficient is 0.49.”

This reflects the findings of a study from last year, a meta-study of the existing literature on the relationship between household carbon footprints and different incomes and expenditures – the income elasticity and expenditure elasticity of emissions, to use the jargon. As this chart shows, while footprints increase they do not match increases in expenditure, let alone the steepness of the increases in income.

The study suggests that the elasticity measures more nearly approach 1 – in other words, the steepness of the increase in the emissions slope more closely matches the income and expenditure increases – in countries where electricity and transport have seen carbon reductions. This makes sense: if the baseline economic activities that are largely common across income groups – heating and lighting housing, and the basics of transport – have been largely decarbonised, the emissions footprints of individuals will much more closely mirror their overall expenditure levels.

Even if this does happen as we decarbonise the economy, and the steepness of the slope showing an increase in carbon emissions does begin to match that for expenditure, it will still be a long way short of the increase in incomes. The unfairnesses in the income distributions will continue to outstrip the unfairnesses in emissions. This strongly suggests that the richest should have excess money to be able readily to afford additional payments to reflect their greater carbon footprints. However, like the loss and damage agreement for nation states, we will have to see whether the rich do indeed step up to this particular challenge. A just transition will need them to.

Sea level rise: the most unjust transition

Assessing US consumers’ carbon footprints reveals outsized impact of the top 1%, Jared Starr, Craig Nicolson, Michael Ash, Ezra M. Markowitz, Daniel Moran, Ecological Economics 205 (2023) 107698

Our World in Data

The outsized carbon footprints of the super-rich, Beatriz Barros, Richard Wilk, Sustainability: Science, Practice and Policy 17 (2021) 316–322

Survival of the Richest, Oxfam America, January 2023

Doughnut Economics Action Lab

Loss and Damage agreement, UN FCC, November 2022

Expenditure elasticity and income elasticity of GHG emissions – a survey of literature on household carbon footprint, Antonin Pottier, Ecological Economics 192 (2022) 107251

But is it FAIR?

Information technology often, sadly, fails to be fair. Some in the sector, though, are trying – and even occasionally beginning to gain traction.

It’s what makes the FAIR Principles (sometimes called the FAIR Data Principles) so interesting. The abbreviation stands for Findability, Accessibility, Interoperability and Reusability, and it’s a set of standards for datasets that are used in analytics. If data lives up to these principles, it is more likely to be capable of reuse and thereby enable others to test if results are replicable – one of the foundations of the scientific method. Allowing others to reuse data and to attempt to replicate results may help remove some errors from new technology – including creating or perpetuating unfairnesses. At present for example, there is a tendency to ossify past unfairness through the biases embedded in the data that is used to train artificial intelligence.

The FAIR principles were launched in a 2016 paper. This makes clear that the ambition for the principles extends beyond narrow definitions:

“Importantly, it is our intent that the principles apply not only to ‘data’ in the conventional sense, but also to the algorithms, tools, and workflows that led to that data.”

The creators of the FAIR Principles appear to have a touching confidence in the ability of the human mind to deal with data in context and to understand nuance, which feels a little like overconfidence to me. But the point of the principles is the recognition that modern techniques rely on machines (they use the term to cover the wide spectrum of technologies that might be applied) to deal with the vast quantities of data that are available, and those machines are usually even less good than humans at grasping context and nuance. The FAIR Principles won’t change that, but they provide enough visibility for other scientists to probe and test the extent of the errors that will have arisen because of those failures, including errors of unfairness.

While the FAIR Principles are not explicitly expected to be fair, their nature and application should lead to fairness as well as FAIRness. ‘Findable’ requires data to have rich metadata attached, which needs also to be ‘Accessible’ by enabling universal free and fair access; ‘Interoperable’ and ‘Reusable’ mean that the data is in a form that can be widely understood and applied and put to use by others without inappropriate restrictions. It’s a level playing field for data availability and use – ensuring that fair access can enable ongoing checking and testing.

Others have since taken the concept further, such as the Dutch moves to create a data stewardship profession that reflects the FAIR Principles. This profession would take responsibility for the shared endeavour that good data stewardship requires, across time and involving individual researchers, other scientists in the research project, their sponsoring body or institute, and indeed from the funders of the research. Recent presentations as part of this project, including ones dated October and November of this year, suggest strong progress in acceptance of the Principles (at least in the healthcare data sector that is the project’s initial focus) but also that there is much more to be done.

A further development has been a move towards automated testing of the FAIRness of datasets. This aims to avoid the significant workload implied by the prior manual approaches to such assessments. The range of 17 core metrics developed by the team helps mitigate against the risk of a single measure being applied to a complex area, which would lead to less insight being offered – as well as the risk of gaming. Each core metric has between one and three practical tests that can be applied to test whether it is met or not.

The initial test, on some common datasets, wasn’t encouraging, with around average expected scoring on findability and interoperability, and significantly below average scores for accessibility and reusability (though the academics note that the particularly poor accessibility result is based just on a single metric, which every dataset failed – they were only applying 13 out of the total 17 metrics at this stage):

The good news is that after engagement with the data repositories and subsequent relatively straightforward improvements to the metadata, the results improved notably (though the result for reusability oddly seems to show a little deterioration in quality by a fifth of the data at the top end):

Applying transparency and accessibility standards more broadly in the data and technology world is of growing importance. Citizens often feel disempowered and disenfranchised in the face of the technology that is all around us. Making tech, including the underlying algorithms that shape our modern experience of the world, more transparent is a necessary step to build confidence and to prevent democratic backing for technology further eroding.

This matters. The algorithm is the echo chamber. It ushers and tempts us from the babble of the town square down narrower side-streets, sometimes forcing us into the most squalid and awful thoroughfares with a one-way traffic of angry bile. Unless we have sufficient transparency to understand how algorithms usher and force people in these ways we cannot address and reverse the harms that they are causing. Unfairness makes people more prone to accept conspiracy thinking, and at present social media algorithms are feeding and exaggerating that process. FAIRness may help fairness over time.

Learning from the stochastic parrots

This blogpost was originally written for Data Ethics Club and also appears among its papers.

The FAIR Guiding Principles for scientific data management and stewardship, Mark Wilkinson et al, Nature Scientific Data, March 2016

Professionalising data stewardship: Dutch projects

Towards FAIR Data Steward as profession for the Lifesciences, Salome Scholtens et al, ZonMw/Zilveren Kruis, October 2019

FAIR data stewardship: the need for capacity building & the role of communities, Mijke Jetten, E4DS Training, October 2022

An automated solution for measuring the progress toward FAIR research data, Anusuriya Devaraju, Robert Huber, Patterns 2, 100370, November 2021

The Gini in the executive pay bottle

What’s the best way to assess the fairness – or unfairness – of executive pay? It’s a lively debate around the world. It has a particular bite in countries where overall inequalities are especially pronounced.

Usually that fairness is assessed, or at least made manifest, through disclosure of pay ratios. The US, France and the UK now both require disclosures by public companies of the ratio between the CEO’s total pay and that of the median worker* (the US requires the comparison to be across the global employee base, with some limited exclusions permitted, the UK only considers the UK employees; neither considers the circumstances of the workforce who are not employees). Campaigners and legislators have focused on this ratio: see, for example, San Francisco’s Overpaid Executive Gross Receipts Tax, which imposes additional taxation on companies that pay their CEOs more than 100 times their median employee; and the Max 12:1 campaign. But there are broader measures that could be considered.

For various reasons I have read a few South African annual reports of late, for the first time for some years. And I was very struck by some of the disclosures made by miner Impala Platinum (usually called Implats for short). There is an explicit section of its latest Annual Report called ‘Our approach to fair pay’ which also references the company’s Fair Pay Policy Statement. This policy statement includes Guiding Principles, among them Principle 3 on Equity, ‘Reward given to different employees is fair, consistent, and justifiable’:

Principle 3a: All jobs are appropriately graded to reflect required technical knowledge, skills and experience
Principle 3b: Reward policies are designed to enable necessary variation depending on local contextual factors, such as in hardship locations, fragile states or absence/scarcity of necessary skills
Principle 3c: There is a commitment to policies being applied systematically

These fine words also appear to be reflected in practice. The annual report notes recent pay deals both for its refinery workers and its miners, and asserts that “The Implats guaranteed minimum wage for permanent full-time employees remains significantly higher than a ‘living wage’, and in addition our employees are eligible for progressive variable pay arrangements which are generally above the median for the industry.” The pay ratio that it discloses is between the CEO and the lowest paid employee, and is 58:1. While this is far above the 12:1 some campaigners are seeking, many US, French and UK companies report pay ratios far in excess of this – and those are ratios to the pay of the median employee, not to the lowest paid.

But still more unusual are the company’s disclosures of income inequality metrics most often applied by economists, the Gini coefficient and Palma ratio (both explained at the foot of the image):

All of this discussion of fairness in executive pay is a response to the expectations of South Africa’s corporate governance code, the latest update of which – referred to as King IV – was published in 2016. As some of the introductory wording to the new Code states: “An important introduction in King IV is that the remuneration of executive management should be fair and responsible in the context of overall employee remuneration. It should be disclosed how this has been addressed. This acknowledges the need to address the gap between the remuneration of executives and those at the lower end of the pay scale.”

And fairness is in fact included in King IV’s Glossary of Terms:

“Fairness refers to the equitable and reasonable treatment of the sources of value creation, including relationship capital as portrayed by the legitimate and reasonable needs, interests and expectations of material stakeholders of the organisation.”

Putting this into practice, Principle 14 reads “The governing body should ensure that the organisation remunerates fairly, responsibly and transparently so as to promote the achievement of strategic objectives and positive outcomes in the short, medium and long term.” Under it, recommended practice 27 states: “The governing body should approve policy that articulates and gives effect to its direction on fair, reasonable and transparent remuneration.” And one expected disclosure in remuneration reports is: “An explanation of how the policy addresses fair and responsible remuneration for executive management in the context of overall employee remuneration.”

This focus on fairness matters particularly in South Africa, which because of the unfairnesses of its past is one of the most unequal societies in the world. The nation’s Gini coefficient and Palma ratio are still worse according to the OECD than those set out in the Implats annual report, though the calculations are from 2017 rather than the latest years: 0.62 and 6.9 respectively. In both cases, the nation sits at the most unequal spot among those countries for which data is disclosed. It is recognised that business must play a significant role in addressing this unfairness.

Not all South African companies agree though that disclosure of these ratios for their employees is the best way to express their attempts to press for fairness. The most blunt that I found was this from the annual report of retailer Woolworths: “While the Gini coefficient or index is widely considered to be the most scientific and accurate measure of income disparity and many commentators use it as a proxy for fair and responsible remuneration as envisaged by King IV, the committee agreed to focus on strategic initiatives to drive and address fair and responsible pay.” It emphasises that it is not tied to South Africa’s minimum wage, nor even to the concepts of a living wage, but rather is aiming for a ‘just wage’, “informed by many data points, including minimum wage rates, market rates, CPI, and our EVP [Employee Value Proposition – the firm’s overall approach to staff retention and reward] strategy”.

Woolworths also produces a striking waterfall chart showing the impact of its own approach:

I have railed previously about the problems that come from a focus on just one metric, and I’ve suggested that the fact considerations of inequality are often reduced simply to a discussion of the Gini coefficient is narrowing and unhelpful. So I don’t rush to approve of Implats’ disclosure of Gini. But it is very clear that disclosing the Gini coefficient, as well as median pay ratios and pay ratios to the lowest paid employee, and also the Palma ratio, offer a much richer understanding of the spread of pay across the whole organisation than disclosing any one of those numbers alone. The US and UK requirements to disclose pay ratios are positive, but it is becoming harder to say that they go far enough; what they do indicate is that many companies have many more issues of unfairness than both Implats and Woolworths seem to have, notwithstanding the difficult and unfair national context that both of those companies face.

It’s particularly welcome that both companies looked at in a little detail for this post disclose their number of employees and that all of them receive at least the living wage. As I’ve said in a previous blog, I think all companies everywhere should be disclosing these datapoints as the start of baseline information on social factors that can be aggregated across portfolios. The South African remuneration approach pushes companies to think much more about fairness, and companies are rising in a variety of positive ways to this challenge. Other countries have something to learn from King IV’s call for fairness, and companies elsewhere in the world from how South African businesses are responding to that call.

* ‘median worker’ is clearly a shorthand for the worker who is at the middle of the pay distribution (often calculated by shorthand methods). No worker is average.

Fairness – the human lens for addressing our current challenges

The madness, let alone unfairness, of US executive pay

San Francisco’s Overpaid Executive Gross Receipts Tax

Max 12:1 Campaign

Impala Platinum Annual Report 2022

Impala Platinum Fair Pay Policy Statement and Guiding Principles

King IV Corporate Governance Code, Institute of Directors South Africa, 2016

OECD data: income inequality

Woolworths Annual Report 2022

Political fairness: a failure

The latest polls carried out under the banner of Lord Ashcroft have just been published. The title They think it’s all over suggests that the news for the ruling UK Conservative Party is not good. Ashcroft is a former treasurer and deputy chair of the Party – often informally called the Tories – and in recent years has mapped its decline. The subtitle of the report Can the Tories turn it round? shows Ashcroft’s own focus. But really the news from much of the polling is bad for any politician: there is not much eagerness and positivity around any party, or the overall political process.

One result in particular stands out for this blog: responses to the question of whether people believe that either the Conservatives or their rivals in the Labour Party ‘stand for fairness’.

It’s welcome that Ashcroft asks this question. With a cost of living crisis and worrying inequalities all around us this could be, perhaps ought to be, the political assessment of our times. Fairness could be an issue that plays for either end of the political spectrum: it isn’t clearly either a left- or a right-wing issue.

The polling result is terrible news for the Conservatives. They face a 25% deficit on the issue, their second largest gap overall, outdone only by the 36% margin on the question of which party wants to help ordinary people get on in life. Only 4% of voters believe that the Conservatives stand for fairness. The underlying data that Ashcroft also discloses reveals that only 9% of those who voted for the party in 2019 believe that they stand for fairness, and just 17% of those expecting to vote for them at the next election do.

But the poll result is also pretty bad news for Labour. Their lead is impressive, but it more reflects the dire result for the Conservatives than a positive view of Labour. Overall, only 29% of people believe that they stand for fairness. Even among their supporters only a small majority believe that they do: 56% of those who supported them in 2019, and 59% of those likely to vote for them next time around. Under half of every other group of voters thinks the Labour Party stands for fairness.

This is surely a political failure. On a crucial issue of public concern that is central to the problems our nation and the world are facing, neither party is demonstrating real leadership.

It does leave an opportunity: seizing and deploying the language of fairness and expressing it through policies could resonate deeply with voters over the next couple of years. Either party could use the concept as a rallying point, indeed potentially a focal point for their manifesto. Will either rise to that challenge?

They think it’s all over, Lord Ashcroft Polls, November 2022

Underlying data tables