The chart shows how, after a long decline, the income share of "the 1%" started to increase around 1980 so that in the US (though not in UK) the 1% are back where they were in the so-called "gilded age". What explains this decline and then rise in inequality, what future direction will it take?

There are two answers to this question, an optimistic one and a pessimistic one, and the answers given reach much farther than you might think.

The optimistic answer begins with Simon Kuznets in 1955. He conjectured that transition from an agricultural to an industrial society starting in the 19th century at first increased inequality as the greater productivity of industrial workers led to higher wages there. However three factors eventually caused inequality to decline again: productivity caught up in agriculture, income from property declined, and there were "institutional changes that reflect decisions concerning social security and full employment"

The astute reader looking at the chart will notice three problems with this theory. First, where is the data for the 19th century? Second what does Kuznets mean by "institutional changes" and third why did inequality start up again?

Unfortunately, the income data from the World Inequality Database does not reach back before 1900. If you find the right economic historian they will fill in some data-points for you, but whether you will (or should) see increasing inequality is not uncontroversial.

Whatever Kuznets meant by "institutional changes" it has since come to mean redistributive taxation and more and better education. Taxes to fund universal health care, unemployment benefit, and a state pension are pretty much self-explanatory but why education? In a speech at Harvard in 2008 Ben Bernanke explained "In the long term, however, the best way by far to improve economic opportunity and to reduce inequality is to increase the educational attainment and skills of American workers". They can then do the skilled jobs of tomorrow (like?). Anyway, an enlightened state drives inequality downwards.

To the third problem why inequality started to go up again around 1980 the original theory has no answer. But it has been rescued by Branko Milanovic "Global Inequality" (2016). He proposes two Kuznets "waves" of rising and falling inequality. The first was for the transition from agriculture to industry, the second is for the transition from industrial to post-industrial service economy. The dotted wave on the chart illustrates the idea.

That completes the optimistic view: though economic changes naturally drives inequality up, state interventions can help drive it down again.

The pessimistic view can begin with Thomas Piketty's "Capital in the 21st Century" (2014). The natural tendency of capitalism is for inequality to increase. Indeed this increase has the status of an economic law that the return on investment (\(r\)) is greater than the rate of economic growth (\(g\)) so that over time the owners of capital accumulate an ever greater share of economic production, wealth concentrates, and inequality increases. The downswing in inequality on the chart is the result of the shocks of two world wars. These destroyed accumulated wealth both via physical destruction and high estate duty. That process has now come to an end so \(r>g\) can once more operate and inequality increase. As for Kuznets, Piketty calls it a "fairy-tale" issued during the height of the Cold War to spread confidence in the benign operation of capitalism. Perhaps it's not altogether fair to call Piketty a total pessimist as he does think a wealth tax can mitigate inequality. So for some serious pessimism we can turn to Walter Scheidel "The Great Leveler" (2018)

Like Piketty, Scheidel sees the natural direction of inequality as upwards, but his view is not just limited to capitalism. Rather the tendency for inequality to increase and wealth to concentrate is universal in all human societies.

To be sure, the original hand-to-mouth existence of a hunter-gatherer group would not permit the accumulation of many possessions because there were not many possessions to accumulate. Some people would be taller than others, some stronger, some heavier, some better looking, some more admired, some men, some women and so on. People would have been unequal in ways that mattered to them and that they could do little or nothing to change. That inequality mattered - but it was ephemeral, when a dominant group member died their advantages died with them.

Persistent inequality had to wait on two things. First there had to be a surplus produced above basic subsistence that could be shared unequally within a society. Economic development that surplus, it is the tap-root that nourishes inequality.

Second there had to be a state structure to protect the distribution of this surplus, to support the inheritance of accumulated surplus from one generation to the next, and finally, to keep the peace.

I will call the surplus producing population peasants, because for most of history that is what they were. The state actor(s) will be the emperor.

The term "emperor" is deliberate. In the pre-modern world the surplus produced per person was small, but as the population grew larger so did the total surplus. At the largest level - empires - the surplus would be enough to allow a few people to become seriously rich. In general the larger the state, the greater the surplus and the greater the potential for inequality. This surely remains true today: in a globalised world the potential for vast fortunes is unprecedented

There is a third actor: the elite. Elites have no functional role as such, they are identified by their wealth, they are the elite because they are the rich. Members of an elite can have a function of course. In the feudal world, for example, the local baron protected his people from the depredations of other barons, as the male lion in a pride protects his cubs from attacks by other male lions (which will kill them given the chance). But the baron is not a member of the elite because he does this but because he extracts surplus from the peasants and is rich as a result.

Before reading Scheidel I tended to assume that the elite were in league with the emperor to screw the peasants. Not so. There is a perennial tension between the emperor and the elite. The emperor wants a robust class of peasants to feed everyone, to provide soldiers, and to pay his taxes. The elite prefer to accumulate the surplus to delight themselves. Often they were tax collectors and exploited the opportunity to siphon off much more than they should have. They tended also to lend money at interest to peasants in difficulties so, by degrees, turning them into their dependents instead of tax payers. Naturally the emperor is antagonistic to the elite.

Piketty identified the world wars of the 20th century as levellers Scheidel generalises this so that at all times and places it is only massive violent shocks that can bring about "powerful" levelling. He calls these shocks the "Four Horsemen of Levelling" and lists them as mass mobilisation warfare, transformative revolution, systemic state collapse, and lethal pandemic. Nothing else will do, and anyway inequality in time returns.

Apart from Scheidel's vast coverage of human societies across space and time it's his explanation of the dynamics of inequality (sketched above) which is so impressive. Of course he recognises that there are policies of the sort proposed by the optimists that would reduce inequality but as he notes acidly: “serious consideration of the means required to mobilize political majorities for implementing any of this advocacy is conspicuous by its absence”.

Yet there is an inevitable chink in Scheidel's pessimism - inevitable because in setting out the dynamics of inequality you must also provide clues for how it might be moderated. Two examples he gives are instructive in this regard, both involve emperors facing serious crises.

In the Warring States period of Chinese history around 359 BCE when (obviously) there was a lot of war, the state of Qin accomplished a major land reform. Plots were laid out large enough to serve a single household (the grid of footpaths and roads apparently still remains) and peasants established on them. Previous elite owners of the land were expropriated. Thus inequality reduced. To be sure, Scheidel says that "initial gains in levelling when established nobilities were displaced, and land reassigned to farmers may well have been eroded and reversed over time as the wealthy employed strategies of reconcentration". Granted this, nonetheless serious military need led to a temporary reduction in inequality. It also led to what Scheidel calls elite recycling where members of the elite lose their wealth (or worse), though others eventually take their place.

In 214 BCE during Hannibal’s invasion of Italy, running out of citizens to conscript, the senate ordered wealthier citizens to hand over some of their slaves to serve as rowers in the navy. The tax was levied according to the following schedule. Census class: 50,000 asses, 1 slave owed. Census class 100,000 asses, 3 slaves owed. Census class 300,000 asses 5 slaves owed. And for the super rich census class 1,000,000 asses and up … 8 slaves owed. Even when the state's existence was threatened the 1% clung grimly to their wealth and shifted the burden on to weaker shoulders as far as they could. The lesson from these examples is that the emperor must not work through the existing elite, for they will defang his efforts. Elite recycling, not a policy agenda, is key.

The present pandemic is certainly a crisis and the British state has spent massively on financial support for those who have suffered from the various lockdowns. Who will pay for all this, will it be Warring States or Rome? It depends on elite recycling.

Well, the peasants have had it with austerity. Thanks to the much maligned first past the post system, which can deliver decisive majorities on a slim advantage in votes, the emperor has a large majority with a lot of new northern MPs. He also shows signs of being impatient with the elite, some very senior civil servants have been replaced. So some elite recycling, and a fair chance that the burden will be shifted on to the wealthy in the manner of WW2. An evens bet, of a modest sum, on Warring States looks reasonable.

In the US, Biden looks to be without a Senate majority, he appears to be picking his team from the class of people who supported Obama, and anyway the elite is much stronger there. Without elite recycling nothing much will change. So it’s a confident bet on the Rome scenario.

There is, however, another straw blowing in the wind. Globalisation is very much out of fashion, and a retreat towards what Keynes long ago called National Self-sufficiency is in the air. Bill Clinton said in 1993 "Nothing we do, nothing we do in this great capital can change the fact that factories or information can flash across the world, that people can move money around in the blink of an eye". But it is easy to stop capital and factories and (thinking of China) even information "flashing" across the world. In a less global world capital will be weakened and labour will be stronger and this should effect some levelling.

I started by noting that the questions lead farther than you might think. The thing is that Milanovic is a globalist. What he calls "high globalisation" led to a massive levelling between countries and raised millions out of poverty: that is a moral case for globalisation. But globalisation increased inequality within countries, with the US and UK being prime examples of this. They provide a moral case against globalisation.

So the question we are left with is a philosophical one: what is the duty of the emperor of a country X towards the citizens of country Y? There are two answers. One is that the emperor's duties towards members of Y are the same as towards members of X because duties arise simply from the status of being human which does not vary with nationality in any way at all. The other view identifies duties as deriving from and bounded by the social networks we were born into. So the emperor's duties are to members of X, because X provides his network. Neither view is perfect, but they condition one's response to globalisation and hence to inequality.