Despite the number of times I’ve heard this assertion, I’ve yet to read a comprehensive explanation or economic analysis of exactly what tax credits subsidise and what the mechanism is through which they depress wages – let alone any empirical evidence. Yet the continual repetition of this line now looks like it could be leading to bad policy. So it’s worth thinking about the principle, theory and empirical evidence behind this claim.
Principle: does the existence of tax credits imply taxpayer ‘subsidy’ to employers?
Living wage campaigners and higher minimum wage proponents seem to believe that a failure of an employer to pay a worker an hourly wage such that they are able to live comfortably (thus necessitating top-ups through tax credits and other support) is equivalent to a subsidy.
Yet employers generally pay people according to the productivity of their work, not some pre-ordained amount to compensate them for a set standard of living. Suppose a worker has low productivity. Say their output is worth £6.50 per hour to an employer who decides to pay that amount in a competitive market (the exact level of the current NMW). Our worker may still be eligible for tax credits because of their family circumstances or the low income of the household as a whole. The Living Wage in London is judged to be £9.15 per hour. The logic in claiming that the tax credit payments are a subsidy to the employer seems to be that it is the employer’s responsibility to ensure his workers meet the living standards set out by the Living Wage Foundation and that any failure to do means he or she is ‘subsidised’ by taxpayers.
To which my question is: why? Whilst we’d all like to see people living comfortably without the need for taxpayer support, it’s not clear to me why the burden for this social ambition is the responsibility of the employer and/or customers of that employer as opposed to anybody else – especially if the employer is already paying the worker something equivalent to his marginal product. In fact, all you are doing in urging the payment of higher wages in this scenario is encouraging the employer not to hire the worker in the first place.
The only reason that you might believe that tax credits are subsidising employers then is if we are talking about an employer being a monopsony in a market – and having the market power to keep wages below the marginal product of its workers. This was one of the arguments in favour of the minimum wage in the first place. Yet proponents of a much higher minimum wage or a Living Wage have provided no evidence this is the case for low paid workers, and it seems very unlikely indeed that competitive industries such as retail, care homes or pubs are characterised by this degree of market power. In all then, the whole ‘subsidy’ rhetoric makes no sense.
Theory: could tax credits depress wages?
Even if you accept my argument above that it makes no sense that to think in the ‘subsidy’ framework, theoretically, there are still reasons why one might expect tax credits to temper wage increases over time. Not because tax credits ‘subsidise employers’, but because they might increase labour force participation and deter productivity growth.
The first of these is surely a feature of the rationale for working tax credits at least, not a bug – and given that extra workers can be complements to existing workers as well as substitutes (see the immigration debate), whether this effect is significant is doubtful.
A more significant argument has been put forward cogently by Howard Flight and others. Given that the current set-up of tax credits entails many households facing high marginal tax rates over large bands of income, one can imagine that on the margin the tax credit system deters human capital accumulation, and makes it less likely that individuals will accept or strive for promotion.
In effect, a means-tested tax credit with a significant taper rate gives poorer households money but makes it more difficult for poorer families to earn more income, thus meaning that compared with a situation where there were no tax credits, employees may be less inclined to upskill or push for higher wages. This can of course effect firms’ decisions too. They might decide to keep on more low skilled workers than invest in labour saving technology. But note again: this does not show firms being subsidised by tax credits, merely that a side-effect of giving means-tested assistance is to create high marginal tax rates which can disincentivise earning more income and so cause lower wages due to the decisions it engenders.
It should also be noted that this phenomenon occurs right up the wage scale. Depending on how many children they have, and on whether they qualify for the childcare element, families on well above-average earnings can still receive tax credits. There is no difference between the arguments used in relation to tax credits and the minimum wage and tax credits and wages at (say) £35,000. If tax credits suppress productivity, they do so a long way up the income scale. This is an argument for reforming or abolishing tax credits not for lifting the minimum wage.
Evidence: do tax credits depress wage growth through this mechanism? And is a higher minimum wage a substitute for tax credits?
Ultimately then, this looks like an empirical issue: is there evidence that wages have been substantially depressed by the existence or expansion of tax credits? There haven’t been many empirical studies on this. The Resolution Foundation has a brief summary of the literature from page 25 here which gives mixed evidence, but none of these existing studies are really transferable to the current UK setup. Nevertheless, they have sought to examine the broad effects by comparing wage growth both in the bottom half of the income distribution (where tax credits are concentrated) to the top half; and by looking at wage growth of eligible and non-eligible groups in the same (lower) part of the earnings distribution.
They find no evidence of differences in wage growth across the bulk of the income distribution between 1999 and 2008 (though acknowledge that there might be some ripple effects up the lower half of the income distribution due to higher minimum wages). Perhaps more importantly, their examination of wage growth among low-wage parents (who are much more likely to be in receipt of tax credits) vs. typical wage earners, higher-waged parents and low-to-middle earning non-parents suggests no significant differences either.
So, if there is little to no evidence that tax credits have led to slower wage growth, why exactly do we constantly hear that tax credits are subsidising low wages?
People continue to assert that increasing the minimum wage towards the Living Wage will save vast amounts of taxpayer money through lower tax credit payments – implying that a higher minimum wage is somehow a ‘substitute’ to in-work benefits. But both parts of this sentence are questionable. The government’s own studies (p.16) have found that increases in minimum wage rates are only likely to lead to minor improvements in the public finances by the time labour demand and the inflationary effects (higher benefit and pension payments etc) are accounted for.
Perhaps more importantly, and as I keep pointing out, the people earning between the current minimum and living wages are often very different from the people receiving tax credits. Three-fifths of those earning less than a Living Wage are part-time workers. Over two-fifths of workers earning less than the Living Wage are actually in the top half of the household income distribution (i.e. they might be second earners or young people in relatively affluent households). Tax credits on the other hand are much more highly targeted at those on low incomes.
What we can say with absolutely certainty is that those who claim employer subsidies amount to £29bn are wholly misguided, aside from any of the arguments above. Child Tax Credit (CTC) does not come with any work requirements attached. As a result, nearly a third of all tax credit recipient households – 1.4m out of 4.6m – have no adult in paid employment. This alone accounts for £8.1 billion. Many other recipients of in-work tax credits likewise work hours substantially shorter than a full working week – out of the 3.2m in-work tax credit recipient households, 0.7 million work fewer than 24 hours per week, and 1 million more work fewer than 30 hours. For these groups tax credits top up people working less than a full week, not low hourly wages.
Given all this, I wish an interviewer would examine much more critically the assertions we constantly hear that tax credits subsidise employers and low wages – and that higher minimum wages are the best means of mitigating this effect. Perhaps I’m missing something given the theory and evidence I’ve looked at. But surely the burden of proof is on those who are pushing for policy change?
Ryan Bourne is the IEA’s Head of Public Policy.