Saturday, 21 September 2013

Is This a Tax or a Subsidy I See Before Me?

So Edward Samuel Miliband, the leader of the Labour Party, has finally promised to abolish the spare room tax, or should that be the spare room subsidy?  I’m not entirely clear.  Every Labour party politician assiduously refers to this as the spare room tax and every member of the Conservative party assiduously refers to the spare room subsidy. Their various supporters in the press follow suit, so which is it?  In essence, this is about the use of language to bias a debate.

  • If it is a tax being imposed on poor people, then that is more evidence that the people imposing it are heartless Tories.
  • If it is a subsidy being withdrawn, then that is further evidence that people who oppose the policy just want to spend money and would explode the deficit if they got power.

This kind of labelling is an old (and quite transparent) trick, and one that I have already used twice in this post alone (although you may have to know what my own political bias is in order to spot one of them).

Technically speaking, I have more sympathy with the Conservative description of the policy as what is happening is that a subsidy is reduced if the family living in subsidised accommodation is judged to have more than sufficient living space.  I suspect some quick witted Conservatives will start saying “there he goes again making promises he can’t keep: You can’t abolish a tax that doesn’t exist.” 

Although withdrawal of a subsidy can have effects similar to the imposition of a tax.  Indeed, a subsidy is simply a negative tax rate, and so there is a point that if a subsidy is reduced, that is mathematically the same as increasing a tax. This often leads some people to cast a reduction in tax rates, or a refusal to tax a good they think should be taxed as subsidising the good in question.

The problem with using this logic to cast the failure to impose taxes you like as a subsidy and the failure to give out a subsidy you like as a tax is that it is highly vulnerable to reductio ad absurdum.  A friend and I were once discussing an article which referred to the withdrawal of a tax on mortgages (which happened about 20 years ago) as a subsidy for private home owners. At the time we were sitting in Oxford’s economics department which is one of those buildings where all the external walls are made of floor to ceiling windows, so I had to point out that the abolition of the Window tax in 1851 meant we were in the most subsidised building in Oxford.  Indeed, under that view, the Shard of Glass in London is receiving an eye wateringly large subsidy (adjusting the taxes that were charged for inflation).

To go one step further into the absurd, before breakfast, I could claim that books have been heavily subsidised by the government’s failure to tax them at £1,000 per book. After breakfast, I can claim that books have been heavily taxed by the government’s failure to subsidise each book by £1,000. The only thing that has actually changed while I was having breakfast is my reference price, not the policy.

So there must be some reference where we can refer to the good as either being taxed or subsidised.  One extremely convenient one would simply be to look at whether the government is paying out money (in which case we should probably refer to a subsidy) or whether the government is taking in money (in which case we should probably refer to a tax).  The key advantage of this way of labelling things is simplicity, to know whether a policy is a subsidy or a tax, you just need to look at what happens to the government’s budget. Under this convention, the Conservatives are correct and the housing benefit changes constitute a reduction in subsidy rather than an increase in tax.

However there is another way of looking at this which allows us to be more sympathetic to Labour. We could say that a policy is a tax if it raises the price above the market price that would prevail if there were no market imperfections, and that it is a subsidy if it lowers the price below the one that would prevail if there were no market imperfections. One attractive property of this way of labelling policies is that we can refer to the refusal to tax petrol in a manner that takes the price of the carbon externality into account as a subsidy.  In this labelling scheme, any policy that involves any kind of subsidy or tax is per se inefficient. Since the price of housing in the UK is kept artificially high by imperfections in the housing market, this would allow us to cast the benefit changes as a tax and say that Mr. Miliband is correct to talk about abolishing the bedroom tax.

If this labelling scheme appeals to you, then I just have one question for you and Mr. Miliband: So you are going to be taxing those who are ineligible for housing benefit how much?

I thought so. It is probably best if we just stick with the definitions of tax and subsidy that refer to the effect on government spending and income. 

None of this makes any comment on whether reducing housing benefit in this way is the correct policy, it is just about whether we should call it a tax or a cut in subsidy. Now might be a good time to start openly discussing the merits of this particular subsidy and whether it should be reduced or not; and if it should be reduced, is this the right way to go about it?


  1. This comment has been removed by a blog administrator.

  2. Nail on head. But I think the fundamental problem is the absence of an absolute reference point, with which policy measures can be compared. Without such a benchmark, any measure that leaves you with less money in your pocket can legitimately be described as a tax increase, and equally rightly as a reduction in subsidy. A measure that leaves you better off can similarly be called either an increase in subsidy or a tax reduction.

    There are rather pedantic arguments being sustained (e.g. by financial journalist Paul Lewis). However, the choice between the two is not a matter of pedantry, but a matter of one's agenda.

  3. A friend in the Oxford Economics department22 September 2013 at 11:15

    Still don't entirely agree.

    Given that the whole thing is reference-point based, I would propose a system whereby ask whether the thing in question is treated more or less favourably than the set of comparable things in the economy. This has the exception that we might be working at a "top level" of this nesting, in which case, we can look in absolute terms, taking externalities into account.

    For example, VAT. The "top level set" of this nesting is consumer goods. Clearly, in absolute terms, there is a tax here. However, now look at children's clothes (or men's razors). No tax... but given the set they clearly belong to, the way similar goods are treated, that's a subsidy. Similarly, home heating fuel with a 5% VAT rate: subsidy (lots of externalities to count there too -- positive and negative -- so I think men's razors and children's clothes are a better example).

    How about 0 VAT on food? Well, perhaps that one's debatable. Is food an example of a consumer good (in which case it's subsidised) or a separate top level good in its own category (in which case it's just something we don't tax)? Unclear. Fine. The system acknowledges the absence of perfectly crisp lines.

    Manor Road building doesn't pay the window tax.... fine, whyever should it, in our current tax system "buildings with windows" are not a set in general which pay tax. Though if the window tax existed in its old form, but university buildings were exempt, I would say it was a subsidy.

    So what about the spare rooms? Tbh, I think I know how the system's being applied, but I'm not exactly sure.

    So if it's: "you are eligible for subsidised housing! This is the one you're getting! These are its rooms! Oh, you no longer need one of them, now your children have left home? Well, this is still the house you're getting, but now you have to pay x back to the govt!"

    then you have to make a payment to the govt relative to what would ordinarily have been expected, given that you have a social house and can't do anything about the size of it. TAX.

    But if it's: "Your house is bigger than you need, you know, and we have a long waiting list. Can you please move to this smaller house? No? You don't want to? Well we'll end up paying more in housing benefit then, we can't justify subsidising you fully in this bigger house"

    then you should (perhaps) be expected to move, or if not, to cover for yourself the extra subsidy entailed in your not moving. SUBSIDY.

    Another way to put it: lack of social housing? True. Underoccupancy problem making this worse? So get to the root of the problem, and provide options and incentives to more optimally arrange families between houses as their circumstances change. Fair enough. No need to subsidise people in large houses they don't need.

    Impose a charge on people who happen to currently be in suboptimally allocated social housing, suddenly, with no availability, possibility or time to move and so address the actual problem?


  4. Elizabeth, again14 October 2013 at 05:50

    You should really Listen Again to today's World at One on this. There's a new report done by York Uni's Centre for Housing Policy ( -- sadly the report itself doesn't seem to be available just yet. Btw, that's routine for academic research in the news, and *really annoying*. What are the press agents thinking?)

    The headline figures I got were these:
    * 20% of people liable for the [tax / subsisdy-loss] have moved house to avoid it
    * The government's impact assessment only expected 8% to have moved by this point.
    * The reduction in housing cost to the government associated with this downsizing is less than the cost of the [tax / subsidy-loss].
    * The government will therefore make about (?) 30% lower "savings" than they expected.

    So look. From the government's own impact assessment, they expected (and budgeted for) 92% of people to stay put. They intended to take this additional charge from those people, *not* to induce them to move. And since the rent remains static for those who don't move, that money would be coming out of the rest of their household budgets. And the money in question is significantly more than the additional cost to the government of letting these people stay in their larger homes.

    They didn't intend or want to induce more efficient use of housing. They wanted to take the money out of the household budgets of the (relatively) poor, justify it on the grounds that these people are receiving "too much" in other ways (in terms of space) and call it a "saving". I'm pretty shocked, actually.

    And yes, this definitely slots the policy squarely into the "tax" interpretation, in my taxonomy above.