It depends…..

…Specifically, for LibDems it depends whether you are in opposition or government….

I think the issue here is whether you think these should be measured over a lifetime or a point in time.The IFS are right as long as higher savings (whether in property or financial assets) which are zero-rated outweigh the zero-weighting of necessities. But a tax that takes more money off people (the same people perhaps) when they are poor than when they are rich might still be considered regressive, I’d think.

There is an alternative intuitive way of looking at it. Before I read up on this, my intuition was the opposite, that the rich spent a high proportion of their income on houses and private education, which are also not VAT’able. And that this made VAT regressive.

In terms of hard data, what the IFS provide in their briefing notes isn’t enough. The IFS and the Treasury have explained the problems with accessing the impact of VAT on the 10% with the lowest income. With their 2010 reports, there are problems of data accuracy when trying to access the impact of VAT on the 10% with the lowest expenditure, and the IFS have acknowledged this.

Their 2009 report ( http://www.ifs.org.uk/budgets/gb2009/09chap10.pdf ), doesn’t seem to have the same problem, and it would be interesting to know why this is. I’m inclined to be persuaded by the argument in the 2009 report. However, the problems with their 2010 data does undermine my confidence.

Ultimately, unless they pitch into these threads with more answers, I’m probably stuck with that uncertainty. I suppose I could pay for IFS membership and get access to their detailed work. Until I’m willing to do that, if I am to draw conclusions from the IFS reports, it’s going to involve a certain level of trust.

I don’t trust the IFS 100%. They’ve acknowledged in their answers in this set of threads that they make assumptions, and sometimes assumptions which I don’t entirely like. But they are assumptions with understandable justifications.

I have enough confidence in them to believe their conclusions are roughly correct,.and that I think VAT is roughly neutral. This is on the basis that they have provided the most authoritative analysis I’ve read on the subject. But no more confidence than that.

It may or may not be “mildly progressive” if the criterion is the percentage of expenditure – unfortunately the quality of the data is too poor to say for sure. It’s much clearer that it’s regressive if the criterion is the percentage of income.

The question that remains unanswered is why you should prefer the former criterion.

The reason why the IFS is intuitively correct to say that the increase in standard rate VAT to 20% is mildly progressive is that the poorest in society wil logically spend a very high proportion of their income on the following: Food, Childrens clothing, Rent, Gas, Electricity, Water, Council Tax and Public Transport.

Most food and all of the rest are either VAT zero-rated, VAT exempt or 5% VAT rate (Gas and Electricity). So none of the above (apart from a small proportion of food) are subject to any extra VAT. That is the intuitive reason why increasing the standard rate VAT to 20% is mildly progressive.

Of course VAT is a regressive tax. Anyone who says otherwise is talking out of the top of their head. It really isn’t worth a thought more.

“A sales tax taxes consumption, whereas an income tax taxes work. If part of the reason we’re in the current mess is we’re consuming too much for the amount of productive work we’re doing, it makes sense to discourage consumption and encourage work.”

Well, if we spend more than we earn, we incur debt. Personal as well as state debt has of course been a bit of a problem (in fact, the blessed Vince used to be worried much more about personal debt than state debt, before the latter gained prominence and everybody told him how clever he had been to foresee it!) However, people normally look to direct influences like interest rates to control the level of debt, rather than indirect factors like changes in the kind of tax levied. Presumably that’s because changes in the kind of tax levied have too indirect and weak an effect to do anything very useful about debt.

If we suppose on the other hand that the level of personal debt is roughly constant, so that income equals expenditure, we can think about what a switch from income tax to VAT might feel like. You get (say) £10 extra in your weekly pay packet, and your significant other tells you that the bills are going up by £10 per week because of the rise in VAT. Does this make you want to race down the pub and drink away your windfall gains? Does it make you race up to the attic to hide away that £10 under the floorborads instead? i’d say neither – it mostly leaves you pretty much unmoved.

So my conclusion is – There are other things that matter more, when deciding whether to income-tax or VAT-tax, than its possible weak influence of the levels of debt and expenditure.

Well, the difference that you pointed out is that VAT on fuel didn’t go up. So the increase is marginally different from the total in its overall distribution. Does that really make a significant difference? Doubt it!

Sorry to make exactly the same observation again, David, but that is NOT what they said. What they actually said is that “We believe that increasing the standard VAT rate in the current system is mildly progressive.”

The are talking about the increase in standard rate VAT, not VAT as a whole.

Yes indeed, I _should_ have said “There are three quite different measures of “progressiveness” here…”

Maybe it would be safest to take a leaf out of Monty Python’s book and say “Among the IFS’s analytical techniques are such diverse measurements of progressiveness as …”

Well, precisely. That’s why most people used to believe that a neutral tax regime was not good enough. The overall balance of taxation should not simply be a little bit more “progressive” than a flat tax. it should be a lot more “progressive”, in terms of income fraction. Unless the tax regime is strongly “progressive”, on the definitions being used here, it will be unfair to the poor, in that it takes a greater share of their (small) disposable income than it takes from the rich.

So – IFS claim that VAT is mildly progressive relative to the flat tax, Steve argues that IFS are wrong and that VAT is regressive relative to a flat tax. In a sense it doesn’t matter that much which of these views is right. Even if the IFS are right, a VAT increase hurts the poor more than it hurts the rich.

In the IFS’s report on the Browne report “A progressive graduate tax after all?” :

http://www.ifs.org.uk/publications/5307

a discussion is given as to whether the Browne poposals are (a) progressive and (b) equate to a graduate tax. This document was instrumental in shaping the debate surrounding the Browne proposals. However, the definition of progressive they used is not consistent with their definition for progressive/regressive for VAT or the standard definition of progressive/regressive taxation as a proportion of income. If a scientist behaved in this manner they would be laughed at and nobody would take them seriously. This, however, is far more serious as their propaganda has had a huge bearing on the debate on the Browne/tuition fee proposals. If it isn’t deliberate propaganda then they are incompetent. Either way, it is a cause of serious concern that such an organisation should be allowed to influence public debate.

“You say the IFS is a joke. So what is your source of perfect, unbiased fiscal analysis then?”

I don’t have to provide a source of of perfect, unbiased fiscal analysis to prove that the IFS’s presentation of their analysis is seriously flawed. How is that question relevant?

Sorry, but you’re completely missing the point.

There are two quite different measures of “progressiveness” here. They mean quite different things. The question is which is the appropriate one to use when evaluating the effects of the government’s actions. Obviously there’s no “right answer.” It’s essentially a political question, not an economic one

It is not a question that a degree in economics makes you any more able to answer (and it is emphatically _not_ a question about which measure is easier to evaluate given the data we happen to have access to).

The question is, in slightly different terms, whether a tax change with the effect I outlined above – taking a smaller percentage of a poor family’s expenditure, but a larger percentage of its income – is that tax change a good or a bad thing? Unless you have an opinion on that, what’s the point of arguing endlessly about the finer technical points of the IFS analysis?

You say the IFS is a joke. So what is your source of perfect, unbiased fiscal analysis then?

The section I referred to in http://www.ifs.org.uk/budgets/gb2009/09chap10.pdf (pages 4 and 5), discusses this.

The Treasury put it this way:
“It should be noted that the bottom decile contains many households with temporarily low incomes, for whom income based analysis, as opposed to expenditure based analysis, may not give an accurate picture of living conditions. In this decile, around 40 per cent of households contain an adult that is self employed or a student. While some of these households will have permanently low incomes, many will not. In contrast, in the second decile, only around 20 per cent of households contain an adult in one of these groups.”

Any of us can come up with hypothetical exceptions to support any line of argument. The question is, are they typical, and do they significantly distort the analysis? I don’t have the expertise, or access to the necessary data, to authoritatively answer that question. Do you?

On the other hand, both the IFS and the Treasury do have that expertise, and have access to the necessary data. They argue that the kind of hypothetical family that does distort the data is more like the following:

(1) A poor one with an annualised income of £10,000 and an expenditure of £10,000
(2) A richer one (someone who is self-employed, but temporarily not working) with an income of £0 and an expenditure of £20,000

“I agree that the phrase “progressive tax” sounds odd when applied to VAT, but the IFS are using it as a technical term.”

But they’re not really, because they are also applying the same criterion (percentage of expenditure) to other tax and benefit changes that have nothing to do with purchase taxes.

The point is that there are two quite different definitions of progressive/regressive being used – one based on a comparison with expenditure and the other based on a comparison with income.

If you’re pushing percentage of expenditure as more appropriate, the first thing you need to do is to give us some arguments why that should be the case.

I’d suggest starting with a hypothetical case in which we compare two households:
(1) A poor one with an income of £10,000 and an expenditure of £10,000
(2) A rich one with an income of £100,000 and an expenditure of £50,000.
A tax change results in the poor household losing £100 a year and the rich one losing £600.
So in terms of expenditure, the poor one is losing 1% and the rich one 1.2%.
But in terms of income, the poor one is losing 1% and the rich one 0.6%.

Is the tax change progressive or regressive, and why?

Matthew,

I agree that VAT will have a greater impact on the poor, for the reasons you give.

But couldn’t the same argument be made of most tax increases, that they would reduce the disposable income of the poor more than that of the rich?

I agree that the phrase “progressive tax” sounds odd when applied to VAT, but the IFS are using it as a technical term. http://en.wikipedia.org/wiki/Progressive_tax

The debate, about how regressive/progressive VAT is, is important, because it may influence which taxes the government raises. Until I read up on it, I hadn’t properly thought through the fact that many other taxes are far more regressive than VAT.

As to whether I think VAT is a mildly progressive tax or not. I am unsure. In my opinion, the IFS puts a pretty strong case. But to prove it, they need to show that the bottom expenditure decile will lose less in a VAT rise than the higher spending deciles. Unfortunately, expenditure deciles using the Office for National Statistics data seem to be skewed by bad data.

I suspect, if I joined the IFS, and could see the detail of their work, I would be convinced. Because I don’t have that information, and because the data we have available looks uncertain, in order to accept that VAT is progressive, I’d have to take their claim on faith.

I do have a fair amount of confidence in the IFS. There are blogs which criticise them, such as Richard Murphy’s, and at times they overclaim the reliability of their conclusions. But they are widely respected, their work is peer reviewed, so I take what they say seriously.

On the basis of what I have read, such as pages 4 and 5 of http://www.ifs.org.uk/budgets/gb2009/09chap10.pdf , my conclusion is that VAT is roughly neutral, and whether it is a progressive or a regressive tax, the difference is unlikely to be significant.

Are people really going to deny that a flat purchase tax with no exemptions would be regressive?

Their ‘definition’ of progressive is not just inconsistent with everyone else, but it is also lacks consistency between the different reports they produce. Their report on the Browne proposals described the tuition fees as progressive on the basis that those on higher salaries pay more; not pay more as a proportion of income, not pay more as a proportion of expenditure, but just simply pay more. They are a joke.

That, in a nutshell, is what the IFS doesn’t believe. Strange but true.