How Sunak came up with disputed claim about Labour tax plans – and the problems with using Treasury numbers like this
Published: June 6, 2024
THE CONVERSATION
The first televised debate between Rishi Sunak and Keir Starmer was dominated by the prime minister’s claim that every working family in the country would be paying £2,000 more in tax if Labour win the general election.
It was a bold claim, which Sunak said was backed up by “official calculations” from the Treasury. Those calculations appeared in a document put out by the Conservatives which accused Labour of making billions of pounds worth of unfunded spending commitments.
Since the debate on June 4, the £2,000 figure has been subjected to a great deal of scrutiny. And it quickly became clear that it was full of holes.
The sums had not been endorsed by Treasury officials, and were based on projections which vastly increased the cost of Labour’s spending plans while underestimating the potential tax revenue Labour says it plans to raise.
Written by academics, edited by journalists, backed by evidence.Get newsletter
Put simply, the Conservatives said that a Labour government would have much less money than it expects to have, and that its spending projects will cost much more than they think.
Overall, some £20 billion (around a third) of the spending totals identified by the Conservative’s dossier appear to be substantially overstated.
The biggest spending item listed in the document is £18.9 billion for Labour’s “green prosperity plan”. But Labour has said that its new fiscal rules would allow it to borrow for investment, so this sum does not necessarily have to be offset against tax revenue.
Elsewhere, the document claims that Labour’s plans to reduce outsourcing, for things like IT systems, will add £6.5 billion to costs. But it gets to this figure by assuming that “insourcing” services is always 7.5% more expensive that using outside contractors. A recent report by the Institute of Government concluded that there is no definitive evidence that this is the case.
Separately, in costing Labour’s offer of free breakfast clubs in primary schools, the assumption is made that 50% of all children will take this up. It also claims that Labour will fully fund the cost of additional meals and staffing, bringing the total cost to £4.5 billion, which Labour denies.
Then there is the price of Labour’s plan to reform local bus services, which the Conservatives estimate to be £3.6 billion. But the small print explaining this figure reads: “This costing has been done at pace with limited data and therefore the uncertainty and risk of error is high.”
There are some unsurprisingly pessimistic assumptions in the revenue raising figures too. For example, the Conservatives say that raising VAT on private schools will raise substantially less than Labour’s estimate of up to £1.7 billion per year. They also suggest that Labour’s plan for taxing non-doms will raise very little (£100 million per year) after the first year.
Then finally, by choosing to add up these purported costs over four years rather than annually, the writers of the Conservatives’ dossier came up with a figure of £58.9 billion in new spending plans. They then subtracted projected revenues of £20.4 billion over the same period to produce a “black hole” total of £38.5 billion.
This big number was divided by the number of working adults (18 million) in the UK to reach the bill of £2,000 per family. (Some viewers might have assumed this was a yearly total rather than a cost over four years.)
The first televised debate between Rishi Sunak and Keir Starmer was dominated by the prime minister’s claim that every working family in the country would be paying £2,000 more in tax if Labour win the general election.
It was a bold claim, which Sunak said was backed up by “official calculations” from the Treasury. Those calculations appeared in a document put out by the Conservatives which accused Labour of making billions of pounds worth of unfunded spending commitments.
Since the debate on June 4, the £2,000 figure has been subjected to a great deal of scrutiny. And it quickly became clear that it was full of holes.
The sums had not been endorsed by Treasury officials, and were based on projections which vastly increased the cost of Labour’s spending plans while underestimating the potential tax revenue Labour says it plans to raise.
Written by academics, edited by journalists, backed by evidence.Get newsletter
Put simply, the Conservatives said that a Labour government would have much less money than it expects to have, and that its spending projects will cost much more than they think.
Overall, some £20 billion (around a third) of the spending totals identified by the Conservative’s dossier appear to be substantially overstated.
The biggest spending item listed in the document is £18.9 billion for Labour’s “green prosperity plan”. But Labour has said that its new fiscal rules would allow it to borrow for investment, so this sum does not necessarily have to be offset against tax revenue.
Elsewhere, the document claims that Labour’s plans to reduce outsourcing, for things like IT systems, will add £6.5 billion to costs. But it gets to this figure by assuming that “insourcing” services is always 7.5% more expensive that using outside contractors. A recent report by the Institute of Government concluded that there is no definitive evidence that this is the case.
Separately, in costing Labour’s offer of free breakfast clubs in primary schools, the assumption is made that 50% of all children will take this up. It also claims that Labour will fully fund the cost of additional meals and staffing, bringing the total cost to £4.5 billion, which Labour denies.
Then there is the price of Labour’s plan to reform local bus services, which the Conservatives estimate to be £3.6 billion. But the small print explaining this figure reads: “This costing has been done at pace with limited data and therefore the uncertainty and risk of error is high.”
There are some unsurprisingly pessimistic assumptions in the revenue raising figures too. For example, the Conservatives say that raising VAT on private schools will raise substantially less than Labour’s estimate of up to £1.7 billion per year. They also suggest that Labour’s plan for taxing non-doms will raise very little (£100 million per year) after the first year.
Then finally, by choosing to add up these purported costs over four years rather than annually, the writers of the Conservatives’ dossier came up with a figure of £58.9 billion in new spending plans. They then subtracted projected revenues of £20.4 billion over the same period to produce a “black hole” total of £38.5 billion.
This big number was divided by the number of working adults (18 million) in the UK to reach the bill of £2,000 per family. (Some viewers might have assumed this was a yearly total rather than a cost over four years.)
But looking at it on a yearly basis, the supposed gap between extra revenue and extra spending is about £10 billion, which in fact is less than the amount of “headroom” that the next government is projected to have to meet its fiscal target.
And of course, if Rachel Reeves becomes chancellor, she will have the chance to set a budget for the next few years, and decide how much revenue to raise. She would also be carrying out a much delayed spending review that will set the targets for governmental spending over the next three years.
So there will be plenty of scope for trade-offs, both in spending and taxation.
A better system
Meanwhile, as Sunak’s £2,000 figure is investigated by the Office for Statistics Regulation, it’s worth remembering that this government is not the only one to try and predict the cost of the opposition’s policies before a general election. As far back as the 2005 general election, Gordon Brown claimed that there was a £35 billion black hole in Conservative spending plans.
And Labour has now accused the Tories of having a £71 billion black hole of their own, based on the assumption that they will immediately implement the chancellor’s wish to eventually eliminate national insurance (and possibly inheritance tax).
But surely there is a better way to inform the public in way that does not give the party in power a huge advance in crunching the numbers. Lord Gus O'Donnell, the former head of the Treasury as well as the civil service, called opposition costings “one of the grubbiest processes I’ve ever been involved in”.
He explained: “Ministers tell you to produce these costings on some assumptions they give you, which are dodgy assumptions designed to make the policy look as bad as possible.”
In the Netherlands, opposition parties can ask a government agency, the Bureau of Economic Analysis, for an official costing of their own proposals.
In the UK, it has been suggested that this could become part of the role of the OBR (although it would require greater resources and more access to government data). If it was, it might just reduce the fog that often clouds debates over spending – and potentially expose issues in the political plans of all parties that they don’t want to talk about.
And of course, if Rachel Reeves becomes chancellor, she will have the chance to set a budget for the next few years, and decide how much revenue to raise. She would also be carrying out a much delayed spending review that will set the targets for governmental spending over the next three years.
So there will be plenty of scope for trade-offs, both in spending and taxation.
A better system
Meanwhile, as Sunak’s £2,000 figure is investigated by the Office for Statistics Regulation, it’s worth remembering that this government is not the only one to try and predict the cost of the opposition’s policies before a general election. As far back as the 2005 general election, Gordon Brown claimed that there was a £35 billion black hole in Conservative spending plans.
And Labour has now accused the Tories of having a £71 billion black hole of their own, based on the assumption that they will immediately implement the chancellor’s wish to eventually eliminate national insurance (and possibly inheritance tax).
But surely there is a better way to inform the public in way that does not give the party in power a huge advance in crunching the numbers. Lord Gus O'Donnell, the former head of the Treasury as well as the civil service, called opposition costings “one of the grubbiest processes I’ve ever been involved in”.
He explained: “Ministers tell you to produce these costings on some assumptions they give you, which are dodgy assumptions designed to make the policy look as bad as possible.”
In the Netherlands, opposition parties can ask a government agency, the Bureau of Economic Analysis, for an official costing of their own proposals.
In the UK, it has been suggested that this could become part of the role of the OBR (although it would require greater resources and more access to government data). If it was, it might just reduce the fog that often clouds debates over spending – and potentially expose issues in the political plans of all parties that they don’t want to talk about.
Author
Steve Schifferes
Honorary Research Fellow, City Political Economy Research Centre, City, University of London
Honorary Research Fellow, City Political Economy Research Centre, City, University of London
