As far as I can see, the government has two main arguments for its current proposals to cut tax credits. One, which is also popular among some on the left, is that tax credits ‘subsidise low pay’, a formulation which is vague enough to mean a number of different things of varying degrees of wrongness –see Gavin Kelly on the “saloon bar economics” of this argument here and the Adam Smith Institute here. The other, which this article addresses, is simply that expenditure has grown from not very many billions of pounds to quite a lot of billions of pounds since tax credits were first introduced, as set out in the Chancellor’s June Budget speech: “The original Tax Credit system cost £1.1 billion in its first year. This year, that cost has reached £30 billion.”
The numbers here are rubbish, as we’ll see, but the problem with the argument is less to do with the numbers than the logic. Public expenditure programmes can grow for any number of reasons, good or bad. Simply stating that expenditure is higher now than in the past is not even the beginning of a rationale for policy change: if it were, we would be discussing cuts to the state pension (up £35bn since 1999/2000, when tax credits were first introduced) and the NHS (up £65bn). When politicians cite expenditure growth as a rationale for cuts, the obvious question is: what led to the rise in spending? In the current debate, that question rarely seems to be raised.
Let’s get the Chancellor’s numbers out of the way before looking at the drivers of tax credit spending. The figure cited for “the first year” of tax credits is not even in real terms- it is the nominal amount without adjusting for inflation. Worse is the fact that the figure relates only to the first six months of tax credits, as they were introduced halfway through the 1999/2000 financial year. Tim Blackwell provides an excellent account here which further shows that as people were still being moved on to tax credits over that six month period the figure cannot even be taken as giving an accurate picture of six months expenditure under the new system, let alone of a full year.
But it gets worse. Tax credits were introduced as a (more generous) replacement for existing benefits. Over time they successively absorbed Family Credit, Disability Working Allowance and, from 2003/4, the child elements in Income Support and Jobseeker’s Allowance. Clearly the pre-existing benefits absorbed into the tax credit system need to be taken into account in looking at expenditure trends. Now the DWP produces a publication, Benefit expenditure and caseload tables, which does precisely this, giving a reasonably consistent time series for tax credits and the benefits it replaced back to 1979/80 (see chart). This shows real terms expenditure of some £8bn on tax credits and equivalent benefits in 1999/00 and £10.5bn in the next year- which as we have seen was the first full year of tax credits. £10.5bn is what tax credits and the benefits they have replaced cost in the first full year of tax credits. £10.5bn is not £1.1bn.
With that out of the way, let’s consider the drivers of growth using DWP’s consistent time series rather than the apples and pear comparisons used by the chancellor. As the chart shows, rather than being continuous, growth in tax credit and equivalent expenditure took place in three distinct episodes: in 2000/01, the first full year of tax credits; in 2003/4 when the system as we know it today was introduced, and in 2008/9-2009/10 when the recession hit.
In the first two of these episodes, the driver of increased spending was policy choice by governments led by Tony Blair: first to introduce tax credits, then to overhaul the system and make entitlements more generous. The 2003/4 increase in spending was very sizeable- so much so that the tax credit system we have today (for the time being) arguably really dates from 2003/4 rather than 1999/00. Compared with the previous system, the new tax credits involved additional expenditure of over £9bn in the first year (including on benefits which were still in the course of being transferred to tax credits). This is by far the biggest change in spending on tax credits since their introduction, and it was a deliberate choice, mainly motivated (at least on the Labour government’s account) by the objective of reducing child poverty while maintaining work incentives.
This brings out the absurdity of simply citing expenditure growth as a justification for cuts. You can’t reasonably criticise a policy decision to increase spending just on the grounds that spending actually increased: you have to challenge the rationale for the original decision- on the basis that you disagree with the objective, or that the policy was not the most effective way of pursuing it.
The last big rise in spending was with the onset of the 2008/9 recession. No surprise there: what is striking is not so much the increase as the fact that it has hardly been reversed, despite cuts and freezes to tax credits since 2010. The main explanation is falls in real wages, with real hourly earnings as of 2014 back to 2003/4 levels across most of the earnings distribution.
(For a fuller analysis, see Steve Machin’s authoritative account here.)
This incidentally shows the vacuity of the government’s argument that the cuts will simply bring tax credit spending back to its level in 2007/8. Reducing spending now to 2007/8 levels means a much less generous system than existed then, because the system currently has to deal with higher levels of demand.
So there is no mystery about the rise in tax credit expenditure: it was driven by overt distributional choices by Labour governments (the main factor) coupled with falls in real wages since 2007/8. The government’s approach mirrors the factors driving growth. The introduction of the National Living Wage can be seen as tackling the wage issue. But as the IFS have argued, it cannot, as a matter of simple arithmetic, completely offset the cuts to tax credits. Thus, ultimately, the government is making distributional choices, which run in the opposite direction to those made by Labour. It is those distributional choices, rather than trends in spending or lazy allegations about subsidising low pay, which should be at the centre of debate.