It was big news. Or at least it might be. Because as always, the devil is in the detail. And so far there is relatively little detail about Ms Reeves's plans for devolving a share of national tax revenues, including from income tax, to ‘regional leaders'.
What taxes, other than income tax? What share and how much revenue?
What responsibilities will be transferred alongside the revenues, given the plan is to be ‘fiscally neutral'?
Will these new revenues and responsibilities be only for areas with mayors or will regions with other leadership structures be in scope at some stage?
And what mechanisms for addressing risk, volatility and large divergences in revenue performance will be put in place?
The answers to these questions will determine just how big a change this is to the funding under regional control, and the transfer of financial incentives and risks from national to regional government.
But some things are clear from the few lines in Reeves's Mais Lecture.
First, the plan is for tax revenue assignment, not tax devolution. The plan – at least for now – is to share some of the revenues from national taxes directly with England's regional governments, not devolve the powers to vary the rates or rules of those taxes.
The simplest form of assignment is just to allocate a share of national tax revenue to regional or local government as a whole and still allocate to individual mayors or combined authorities on the basis of existing funding formulae. Some have suggested such an approach could bolster the bargaining power of local government: this revenue would be theirs by right, and potentially harder to cut back than ordinary grant funding.
Starting with income tax instead makes a lot of sense. The partial devolution (not just assignment) of income tax to Scotland and Wales shows that HMRC can already allocate revenues from this tax reliably across places. It does this based on the main place of residence of each individual income taxpayer, with a series of well-defined rules in place for people with multiple residences.
The second thing that's clear from Ms Reeves's lecture is that she has something more in mind – local tax revenue assignment. Rather than allocate a share of national tax revenues to regional government as a whole, this entails allocating each combined authority a portion of the estimated revenue raised in its own area. With the amount of revenue received depending directly on the performance of tax revenues in their area, this would give regional leaders stronger financial incentives to take action that boosts the regional economy and hence the regional tax base. It would also mean more risk – as tax bases and revenues can rise and fall for reasons completely outside of regional leaders' control – such as global trade wars or energy price shocks.
I said ‘estimated revenues' above quite deliberately. For many taxes, the revenues raised in an area are not known. For example, many businesses have operations that span multiple combined authority areas. How much of the profits of a supermarket chain relate to its stores, versus its website, versus its warehouses and headquarters functions (such as marketing)? Similarly, how much does each part of its operation contribute to its value-added (effectively, profits plus wage bill)? We don't know and if we wanted to assign corporation tax and VAT on a regional basis, would have to instead estimate regional [KO1.1]revenues. This would likely be difficult – indeed, after a decade of work, a sufficiently reliable method of estimating Scotland's share of VAT revenues to enable the assignment of half to the Scottish Government has proved elusive.
Starting with income tax instead makes a lot of sense. The partial devolution (not just assignment) of income tax to Scotland and Wales shows that HMRC can already allocate revenues from this tax reliably across places. It does this based on the main place of residence of each individual income taxpayer, with a series of well-defined rules in place for people with multiple residences.
Because the revenues assigned would depend on the tax paid by the people living rather than just working in a combined authority area, it would provide a different set of financial incentives to current business rates retention. Currently, councils and (some) combined authorities gain from business rates retention only if additional economic activity means new or improved business property in their area. Benefiting from a share of income tax revenue growth would mean combined authorities benefit from the income of people working from home, or commuting to better jobs in neighbouring areas. It could encourage a bigger focus on investing in the skills of people and connecting them to better jobs, rather than focusing so much on business property development.
Income tax assignment could also be a stepping stone to partial income tax devolution to regional or local government in England – letting combined authorities and/or councils vary the tax rate a little to raise more or less revenue. As myself and Neil Amin-Smith – then a research economist at the Institute for Fiscal Studies (IFS), and now one of Rachel Reeves's advisors – showed in a 2019 report, restricting regional/local income tax powers to a ‘flat tax' above the personal allowance (as in Scandinavia), would help address a range of legitimate concerns people (including myself) have with unconstrained tax devolution.
Whether Ms Reeves has the appetite to push beyond tax assignment to tax devolution is unclear. And there are a range of practical issues still to resolve with tax assignment – not least how to design redistribution and risk-mitigation measures that fit into a wider local government finance system that has just seen major reforms.
But she's fired a rocket booster under a debate that has been ongoing for years in regional and local government circles. And it's a debate that we at the IFS hope to continue to inform and influence.
David Phillips is head of devolved and local government finance at the Institute for Fiscal Studies
