“Real household incomes are about £900 per household lower than we forecast in May 2016, which is a lot of money. The question is why and what drove that difference. Some of it, and we can’t be absolute about it, but some of it is arguably ascribed to Brexit”.
Mark Carney, 22 May, 2018
“The Brexit vote means families are already £900 a year worse off”.
Liz Saville Roberts, 6 June 2018
“Bank chief: families £900 worse off thanks to Brexit”
Evening Standard, 22 May 2018
“Brexit vote has cost each UK household £900, says Mark Carney”
Guardian, 22 May 2018
“Brexit costs households £900 a year, says Carney”
The Times, 23 May 2018
The governor of the Bank of England, Mark Carney, told the Treasury Select Committee that real household incomes were around £900 lower than had been forecast before the EU referendum. He said arguably some of this change could be put down to Brexit—this was widely misreported as the governor saying the change was entirely and definitely due to Brexit.
The data behind the Bank’s calculations isn’t publicly available, but it told us the figures it used and how the calculation was done, which we’ll go into later. Using publicly available data we were able to roughly recreate the calculation, but not exactly.
The Bank of England looked at the difference between its projection for real average wage growth (made before the EU referendum) and how much wages actually grew by the first quarter of 2018.
Accounting for inflation over that time, the Bank found that average weekly wages were 4% lower by the start of 2018 than projected. Applying that reduction to average annual salaries in 2016, the Bank concluded that average wages per worker were around £900 to £1,000 lower than projected. This amount wasn’t per household, as Mr Carney said.
Forecasts aren’t exact predictions of the future, and we don’t know what would have happened under different circumstances, for example if the UK had voted to remain in the EU. Without being able to see the exact data the Bank of England used and compare it with similar forecasts we can’t say how much of the £900 is simply down to the fact it’s comparing a forecast with reality.
We think that the evidence behind these sorts of figures should be published and open to public scrutiny, particularly given the high levels of front page media reporting.
Honesty in public debate matters
You can help us take action – and get our regular free email
How the £900 figure was calculated
The Bank of England told us the figure was calculated by comparing real wage growth from just after the EU referendum until the first three months of 2018, with projections it made for that period back in 2016 before the referendum took place.
The exact figures the Bank used to do the calculation aren’t publically available and it hasn’t published its analysis.
It’s not possible to exactly recreate the Bank’s calculation with published data partly because it calculated the change cumulatively over seven quarters. Because most of the published data looks at annual change, only a rough approximation can be reached.
The Bank told us its unpublished data projected that Consumer Price Inflation (CPI—a measure of inflation) would grow by 3%, and average weekly wages by 6.5% over the seven quarters it looked at.
Based on this, it projected that real wage growth would be around 3.5% between mid-2016 and early 2018.
The Bank told us its data showed average weekly wages actually grew by 4% and CPI grew by around 4.75% over these seven quarters. Based on this, it says real wages actually shrank by 0.75%.
The Bank calculated that real wage growth was 4% lower than it had projected would be the case back in 2016. In 2016 median gross annual pay for employee jobs in the UK, was just over £23,000—4% of this is £920, its estimate for how much less the average worker has in their pockets since the referendum.
This figure refers to individual workers, not households as Mr Carney said. If you use the Bank’s figures to calculate the difference, then real wage growth was around 4.25% lower than predicted, which is closer to £1,000.
Using publicly available data we were able to roughly approximate the figures that the Bank calculated, but we weren’t able to replicate them exactly without seeing the Bank’s data.
The Bank’s own Statistical Code of Practice states that ‘statistical methods and practices will be open to public scrutiny’. It’s disappointing that Mr Carney used figures that aren’t publicly available in his appearance before the Treasury Select Committee, particularly given the high amount of media attention the claim received afterwards. The Bank did engage with us over a number of emails and telephone calls to explain their calculations, but figures like these should be published in full so that anyone can check where they’re from and how they’re calculated.
Did Mark Carney say it was all down to Brexit?
Several media outlets reported that Dr Carney had put the £900 loss down to Brexit.
After mentioning the £900 figure to the Treasury Select Committee, Mr Carney explained what he thought was behind the change. He said: “The question is why and what drove that difference.”
“Some of it, and we can’t be absolute about it, but some of it is arguably ascribed to Brexit.”
Mr Carney suggests the link with Brexit is uncertain. We’ve looked in more detail at how much forecasts can change over time and how they eventually compare to reality here.