Sir Keir Starmer is facing demands to make reparations for slavery at a meeting of Commonwealth countries.

The most widely used figure is that the UK owes £18 trillion to the ancestors of slaves – an amount worth seven times Britain’s annual economic output.

So who do we owe this to – and how did they come up with such an alarming figure?

Who do we owe the money to?

The top line figure comes from the Brattle Group Report on Reparations for Transatlantic Chattel Slavery published last June.

The report concludes that combined, western countries owe the descendants of slaves £100 trillion in reparations for both the slavery and post-emancipation periods.

Of this, the UK owes £20 trillion, made up mostly of the £18 trillion in compensation for the period of enslavement.

This apparently makes the UK the second most indebted country, after the United States (£29 trillion). Portugal and Spain come in close third and fourth with £18 trillion owed each for both pre and post-slavery periods.

Most of the British money would go to Jamaica and Barbados, where the report estimates almost two million slaves were transported to, were born or died.

How do they come to the figure?

The figure for the slavery period is derived from five sources.

Around half of the total figure comes from lost wages – some £43 trillion across all western countries.

To reach the figure, they calculated lost wages compared to a non-slave in similar roles, for the average lifespan of a slave. Unlike other workers, slaves were not paid. The researchers then added presumed wages if that slave had lived as long as the average non-slave, because they died younger.

The report does not deduct food and accommodation from the figure, as some previous estimates have. It does however stick to just a twelve-hour working day – the average for manual labourers of the time – rather than assuming slaves worked 24 hours a day as some other sources do.

For the 1.2 million slaves who died in transit, the academics simply applied a “lost-earnings” calculation across the whole estimated lifespan if they had lived.

The rest of the figures came from loss of liberty (£8.9 trillion), personal injury (£4.8 trillion), mental pain and anguish (£18.7 trillion) and gender-based violence (£9.2 trillion).

What are the problems with the analysis?

The main issue, cited often in the report, is a lack of “direct historical estimates” of compensation throughout the period. This means that many of the calculations are based on current payments adjusted for slavery victims with broad assumptions.

For example, the “personal injury” payment is based on the fact that personal injury victims of the 9/11 attacks received one tenth of the amount paid to those who died. The researchers then assume that all slaves faced personal injury, and add a tenth to the “loss of earnings” bill.

Similarly, the payments to the estimated 2.9 million slavery victims of sexual violence takes an average of court-mandated payments from the past 30 years – coming to $50,000 (£38,500) per enslaved woman, per year of bondage.

On mental pain, the report takes a $30,000 (£23,100) compensation figure from the United Nations. It then estimates that slaves lived in bondage for a combined 800 million years, and multiplies that by the contemporary UN figure.

The report doesn’t really acknowledge that the reason there are no contemporary comparisons for injury, sexual violence or mental harm is that in the pre-human rights, pre-welfare state era very few had such protections for slaves or non-slaves. Free people would not have been entitled to any compensation either.

Other harms, such as lost wages, are measured in nominal terms at the time they occurred, and then brought forward to current values using an interest rate.

Given the historically large time scales involved, a significant portion of the final bill is the result of accumulated interest.

The report seeks to both “compensate for loss of purchasing power” and apply a “conservative measure of time value of money” – the basic economic concept which says money in the present is worth more than in the future.

This is so as to not “implicitly allow the enslaving countries to benefit from the delay in recognising the amount of reparations owed”.

The rate for loss of purchasing power is taken to be 1.5 per cent, borrowed from the academic literature, to which 1 per cent is added for the time value of money. Together, this produces a rate of 2.5 per cent.

An alternative 2.3 per cent is also applied in places, representing the estimated average growth rate in wages since the enslavement period.

Could the government afford this?

The report comes with a useful payment plan for the British government.

If the UK was to pay off the reparations for the next 25 years (with an interest rate of 2.5 per cent), it would be roughly £740 billion pounds a year.

In context, that would require the government finding a pot of money four times the NHS budget each year to meet the payments.

Country’s can, in times of desperation, find vast amounts of money through borrowing or increased taxation. The most recent comparison would be Covid, which cost up to £410 billion in additional spending, but even that is just half the amount needed each year.

However, the authors’ reports correctly highlight that reparations are not without precedence.

Germany only paid off its First World War payments in 2010. In 2016, half a billion dollars was paid to Native-American tribes for historic wrongs, whilst $1.2 billion was paid to interned Japanese-Americans after the Second World War.

And of course, some £8.3 billion was paid to British slave-owners for the financial inconvenience they faced at abolition, the report says.

The slavery report calls for 2,250 times that today, an impossibly high number – and given some of the flaws in the analysis, one only meant as a starting point.