Wednesday, 5 October 2011

VAT Incidence Fun

Tesco's results for the six months to August 2011 are out, so we can now compare and contrast four periods in which the main VAT rate was 17.5%, 15%, 17.5% and 20% respectively:

Six months to August 2008, VAT 17.5%
Gross sales £28.1 bn, net sales £25.6 bn, trading profit £1,367 m = profit margin 5.3%

Six months to August 2009, VAT 15%
Gross sales £30.4 bn, net sales £27.8 bn, trading profit £1,551 m = profit margin 5.6%.

Six months to August 2010, VAT 17.5%
Gross sales £32.6 bn, net sales £29.5, trading profit £1,710 m = profit margin 5.8%.

Six months to August 2011, VAT 20%
Gross sales £35.5 bn, net sales £31.8 bn, trading profit £1,773 = profit margin 5.6%.

Ho hum, that fits in with my observation that VAT is partly - or even largely - borne by the supplier. There are four ways of looking at this:

i. If it were true that 'the consumer pays the tax', then we'd expect trading profit to be fairly stable as a % of net sales, and to fluctuate rather more as a % of gross sales. If (as common sense suggests), the consumer has a fixed budget and the supplier bears a large chunk of the VAT, we'd expect the opposite. Trading profits as a % of net sales (as shown above) fluctuated between 5.3% and 5.8%, but trading profits as a % of gross sales fluctuated between 4.9% and 5.2%, i.e. rather less.

ii. Even though the recession started in 2008, Tesco managed to improve their gross profit margin in the next year (because VAT went down to 15%) and after VAT was increased to 20%, their margin went down, as we'd expect. The figure that throws this is the increase in gross margin between 09 and 10, even though VAT went back up to 17.5%.

iii. Look at how customers shifted their spending between non VAT-able items (basic food) and VAT-able items (everything else). Back in 2008, when VAT was 17.5% (and had been for a very long time), 55.8% of Tesco's sales were of VAT-able items. When VAT went down to 15%, this went up quite markedly to 62.4%, then it went down to 60% and then to 58.2% (all as you'd expect).

iv. If VAT were entirely borne by the consumer, the trading profit + VAT ('TP+VAT') would increase/fall when VAT increases/falls; if VAT were borne entirely by the supplier, then TP+VAT would stay the same. Let's assume that 60% of Tesco's sales are VAT-able (from iii.), so a 2.5% VAT change = 1.5% of sales.

In 2008, TP+VAT = 13.8% of turnover, that's our starting point.

In 2009, VAT went down, so if the supplier bears the tax he also benefits from a reduction and TP+VAT would stay at 13.8%. If the consumer benefited from the cut, TP+VAT would go down to 12.3%. As it happens, TP+VAT fell slightly to 13.7%, so the supplier kept 14/15 of the tax cut.

In 2010, VAT went up again. TP+VAT went up to 14.8%, not to [13.7% + 1.5% = 15.2%], so the supplier bore 4/15 of the tax hike.

In 2011, VAT went up yet again, and TP+VAT went up to 15.4%, not to [14.8% + 1.5% = 16.3%], so the supplier bore 9/15 of the tax hike.

We can average out 14/15, 4/15 and 9/15 to 9/15 = 60%, i.e. in summary, my rule of thumb that suppliers bear two-thirds of the VAT seems to be fairly reliable.


Anonymous said...

Well it's like taxing high earners lowers the pay of lower earners they would employ.


QP said...

I don't doubt what you're saying but I would like to see the figures from a number of other retailers before confidently concluding this is the effect we see in those numbers. Would be interesting to see.

Mark Wadsworth said...

AC1, that is a similar effect.

QP, I focus on supermarkets because they actually publish the gross and net of VAT figures; we all understand them; and because they sell a fair cross section of all manner of goods (some with price elastic and some with price inelastic demand).

Whether the supplier bears a third, half or two-thirds of the VAT is all moot anyway, because it's people doing the bearing (employers, shareholders, suppliers to suppliers) so VAT acts to choke off economic activity (and hence receipts from less bad taxes such as Business Rates or income/corporation tax) on a grand scale.

You can do the same exercise for the other supermarkets and you'll get much the same result. The problem is, there's a survivorship bias - we no longer know what the figures would have been for MFI, Woolies and so on who were driven out of business by VAT and/or landlords refusing to drop rents.

QP said...

Ok, thought experiment:
What would happen if Tesco was allowed to put everyone else out of business and effectively became a monopoly retailer of groceries. What would VAT become then, would it be a "less bad" tax because it would be the state sharing in Tesco's newly achieved monopoly profits?

Robin Smith said...

Enployers and employees pay VAT. Both are both consumers and producers.

Its a false battle between labour and capital you are fighting.

The real battle is between labour and capital on one side. And landowners and banks on the other.

VAT is a tax that protects the earnings of land from tax. Rent.

Its as simple as that. Yours is splendid work nonetheless.

Dr. Adrian Wrigley said...

I looked into a similar question. "do changes in VAT rate feed into RPI?". Unfortunately, the UK data going back to 1974 aren't absolutely unambiguous. But it would appear that VAT changes have the "obvious" effect on RPI for several months, after which prices go back to where they were.

So if VAT is raised from 8% to 15%, shops would put up prices sharply, then mostly find they have to drop them again. Somebody's profits are hit, and this will lead to reduced competition over time.

So like most other taxes in use, it is redundant (ie ineffective) as a source of government revenue because it simply takes from things like Corporation Tax, Income Tax, Business Rates and mortgage payments.

Another analysis I did was examining prices at online retailers just before and just after the latest increase in VAT. Again it isn't absolutely clear, but it looks like own-brands rose in price while top brands didn't. I compared Duracell(tm) vs Tesco batteries, Domestos(tm) bleach. Also commodity computer hardware rose but video games didn't. Supports the "taxation comes out of rents" viewpoint.

In France, VAT on restaurant food was recently cut from 19.6% to 5% IIRC. The promotional stickers declared that lower VAT meant lower prices. Actually prices barely budged (but many marginal restaurants were "saved").

I think the situation is fairly clear. But there's endless amounts of tosh written by economists, politicians and econobloggers suggesting genuinely don't get it!

Robin Smith said...

Tax comes out of rent thresholds

Mark Wadsworth said...

AW, ta for extra info.

The effect is most easily observed when there is a rapid large fall or rise in the VAT rate, such as restaurants in France (or Ireland), if the rate is cut from 19.6% to 5%, we would expect to see retail prices fall by about 5% (a third of the cut).

RS, yes, there is a school of thought that says "all taxes come out of rents" which is a fair approximation, but if those taxes are then spent on things which push up rents again ..?

Mark Wadsworth said...

QP, good one.

If there is a monopoly supplier (let's say it's a natural monopoly, i.e. the minimum efficient scale is more than half the size of the total market), then the government can either

a) Just cap prices (like with water). This has the advantage that it seems to work and everybody can afford mains water.

b) Let them charge as much as they like and leave them in peace (as they do with land, to a large extent). This has the advantage that it conserves use of scarce resources because equilibrium quantity consumed is much lower than with (a) where water is wasted (whereby it's not the physical water, its the effort and energy put into purifying and transporting it that matters).

c) All the monopolist to set prices as he likes (same benefit as (b)) and then tax him on the monopoly element, and then hopefully dish this out as a Citizen's Dividend (to enable people to afford it, sane as (a)). This is my favoured approach, of course.

Robin Smith said...

MW It would create a new higher equilibrium. But to landowners where land was property. And to the earnings of labour and capital where it was common property.

If land is free, that is, there is no rent, tax will come out of the earnings of labour and capital. Where else?

If there is rent, that is, there is no free land left, and where land is common property with say lvt, tax will come fully out of rent, of course. Tax is rent by definition.

But where land is private property, then tax will still come out of rent AND both will eat into the customary earnings of work and enterprise.

If we abolished a bad tax the surplus would go back into rent alone.

This is what forces earned incomes down as rent rises. Where else can it come from?

This is not a school of thought. It is a fact. Look and you will see it.

Mark Wadsworth said...

RS, I'm not disputing the basic facts, I was just expressing the same idea a bit differently.