The Laffer Curve, again.

Interesting linguistic point first, to lighten the mood: no European language has a proper word for "tax", because the concept of having to hand over X per cent of your earnings or output or wealth to the government every year is a relatively new and unnatural one. So each language just uses a word at random.

For example:
English "tax" actually means estimate and "duty" means what it means,
German "Steuer" is not actually derived from their word for steer (as in steering wheel), it is derived from an old word for support or prop,
French "impôt" and Italian "imposta" mean imposition,
Dutch "belasting" means burden,
Danish "skat" and Norwegian "skatt" mean treasure.
---------------------------
Right, down to business. The Scottish government can now, in theory, set its own income tax rates, i.e. abolish it by setting a rate of zero. Somebody who's campaigning for LVT in Scotland (to replace income tax) asked me whether I knew of any official or semi-official estimates of the deadweight cost of taxes.

He himself got a curt reply from HM Treasury years ago saying they thought is was about 30p for ever £1 (thirty per cent) of the amount raised in tax. To my mind, this is good start. The overall average marginal tax rate on income and output (taking income tax, National Insurance, VAT, corporation tax and working tax credit withdrawal into account) is around fifty per cent. Thirty per cent of fifty per cent is fifteen per cent, so that's the bare minimum amount by which the size of the economy is depressed (chances are it is much higher, especially if you factor in faster future growth). So if Scotland were to scrap income tax, that would reduce the overall average rate to about 35%.

The only official Treasury nod towards the existence of deadweight costs and the Laffer Curve was a couple of years ago, which I posted about at the time. The point being that the Laffer Curve looks specifically at tax revenues, but the fact there is a curve tells us that the economy is more depressed the higher tax rates are; obviously, with a 100% tax rate, revenues are nil and the size of the economy is zero.

We can turn their calculations into a chart and then work backwards. I shall assume that they are not completely stupid and that 50% is indeed the revenue maximising tax rate:


You can work out implied GDP by multiplying up tax rate by yield. Not all of the fall in implied GDP is down to the economy contracting, say HM Treasury, 60% of the apparent fall is down to evasion and tax planning and 40% is down to the economy contracting.

So at a tax rate of 50%, tax revenues are £31 and implied GDP is £62. At a tax rate of 35% (the lowest which Scotland can have), tax revenues are £28 and implied GDP is £80. The actual increase in GDP is not from £62 to £80 though, (a 30% increase) it is 'only' 40% of that i.e. 12%.

So there's your answer: if Scotland got rid of income tax, its economy would grow by twelve per cent. That's a significant amount. If it collects as much in (new) LVT as it would have done in (old) income tax, its citizens still end up 8% better off (35% of that 12% is lost in VAT and NIC).

Economic Myths: The Fiscal Multiplier

From City AM Forum:

In early 2013, Blanchard and his colleague Daniel Leigh published an IMF working paper on the size of the fiscal multiplier. The multiplier, a theoretical concept invented by John Maynard Keynes in the 1930s, is the most fundamental concept in the whole of macroeconomics.

It measures the eventual impact on the economy as a whole, GDP, of a sustained increase or decrease in public spending. An increase in such expenditure brings more people into work, they in turn will have more to spend, the companies whose products they buy will have more revenue, and will employ even more people. The initial impact is multiplied through the economy.

Sounds simple. But there are many potentially offsetting factors to take into account. Some extra spending will be on imports, for example, which does not boost domestic output at all. The bigger public deficit which the extra spending creates may lead to higher interest rates.

Economists have struggled for decades to arrive at a consensus on how big the multiplier really is. While still being far from agreement, there is a general view that it is low. Indeed, a fiscal expansion, once all the other feedbacks are taken into account, may even lead to GDP rising by less than the size of the stimulus.

In contrast, Blanchard and Leigh argued that, in the current circumstances, it is large and positive. So a fiscal contraction, the basis of the chancellor’s policies, will lead to the opposite, to a sharp reduction in GDP. Events have shown this to be wrong.

They are truly mental.

This is just political bullshit. The lefties always say that the fiscal multiplier is greater than unity (which can't possibly be true in all cases) and the right wingers always say that the fiscal multiplier is less than unity (which can't possibly be true in all cases either).

Allow me to explain...

There are lots of ways a government can finance spending - by increasing any one of dozens of taxes or by running a deficit, which could be financed long term, short term, by 'printing money' etc.

There are also nigh infinite ways a government can spend money [long, long list] and one of the ways in which it can 'spend' money is by just giving people money back, whether as subsidies, tax cuts or higher welfare payments - and the categories all overlap. For example, are Working Tax Credits a subsidy to low paying employers; a tax cut for low paid workers; or a welfare payment?

So there are in turn more or less infinite comparisons you can make by choosing one source of finance and matching it with one type of spending. Some of these combinations will be hugely positive and some will be hugely negative.

For example: increasing taxes on output employment and spending it on Working Tax Credits is nigh self-defeating and a huge negative. Conversely, increasing taxes on land values and spending it by cutting taxes on output and employment is a large positive.

And raising a modest amount in any taxes, i.e. up to 5% or 10% of GDP (however damaging the taxes might be in isolation) and spending it on the core functions of the state: law and order, defence, roads, public health and immigration control is always a massive, huge great positive (notwithstanding that the gains might not be shared equally or even fairly).

It would be impossible to chuck all these infinite possible combinations into a pot and average them out to anything within a margin of error of thousands of per cent.

So it makes much more sense to imagine a Laffer Curve for the spending side. The government just pays for the core functions first (massive, huge net gain), then spends on stuff with smaller and smaller net gains until it reaches the top of the spending curve where the 'multiplier' is unity, and then it stops.

Similarly, there is a Laffer Curve for different types of taxes, there are good taxes and bad taxes. Once a government has collected as much as it can from good taxes it stops. There is no point imposing bad taxes.

With a bit of luck and a tail wind, total revenues at the top of this curve will be exactly enough to finance the spending at the top of the spending curve. Remember that both curves are pretty flat at the top, it is not going to be too difficult to find a level at which they intersect.
-------------------------------------
UPDATE: As PaulC points out in the comments, the whole premise of the current political debate in the UK is wrong. If we take government's word for it that the economy is growing, then the Labour/high spending protagonists have won the argument hands down!

Remeber that despite the Indian Bicycle Marketing narrative, the Lib-Cons did not and have not implemented any sort of 'austerity' programme (except for people they don't like, like unemployed people under 25 etc). Spending was going up even under Plan A and is shooting up under Plan B. Quite what they are spending it on is an absolute mystery to me, probably a load of crap like corporate subsidies.

Which either means that the lefties are right and I am wrong, or possibly that the government is lying and the so-called recovery is built on sand.

Ed Miliband explains the Laffer Curve

From The Telegraph:

Ms Kearney said: "All I am asking you is to explain how you are going to boost growth and that is by spending more and that is by borrowing more if you were Prime Minister at the moment, I just want a straight answer to that."

The Labour leader replied: "Well, I am giving you a very straight answer. I think borrowing would be lower under a Labour government over the medium term."

A temporary one-year VAT cut would cost the Treasury about £12 billion but Mr Miliband could not say exactly how this would be funded in the short term.

"The whole point about a VAT cut is that it would get growth moving and if you get growth moving you get more tax revenues in and therefore you get borrowing down," he said.

The pol's have collectively painted themselves nicely into the corner here.

i. The orthodox-mainstream view is that there is no Laffer Curve. The notion that lower tax rates lead to higher growth is usually only ever vaguely hinted out. The exception is fairly right-wing Tories who point out that reducing rates of tax on earnings and output from "very high" to "a bit lower" can actually boost receipts as well.

ii. But it is also the orthodox-mainstream view across the spectrum and especially among right-wing Tories that VAT is magically a tax on "consumption" not on "production" and therefore has little or no Laffer Curve effects.

iii. For some reason, Ed Balls (for all his faults and they are many), is the only one who has seen through this bullshit and doggedly calls for VAT to be reduced a bit. I'm not sure if he's aware that VAT is borne by the producer not the consumer, perhaps he is? The Two Eds' problem here is explaining that despite what everybody has been saying for years that a) VAT is actually a tax on production (when the rate is cut, prices hardly go down) and b) of course there is a Laffer Curve.

iv. And of course the orthodox-mainstream view is also that taxes on rental-monopoly income streams (like LVT) would somehow have a disastrous effect on the economy and that the Laffer Curve suddenly applies in spades, when quite the reverse is true. If anything, LVT would stimulate the economy (with or without other tax cuts) and there simply is no Laffer Curve, a government can tax rental-monopoly incomes streams at 100% if it likes, as the tax base itself cannot be eroded (if it could be, then it wouldn't be rental-monopoly income by definition). Ah well.

Fun Online Polls: R-Patz, K-Stew & Universal Credit

The responses to last week's Fun Online Poll were as follows:

R-Patz and K-Stew...

On - 1%
Off - 7%
On - 4%
Off - 5%
On - 3%
Who or what are you talking about? - 80%

Well, we were all wrong.

As it turns out, the R-Patz/K-Stew saga took another interesting twist a couple of days after I launched the poll, so maybe they are still on their third "off" or maybe they are already on their fourth "on".

Who knows? Who cares? Kids, bless 'em, is all I can say. The whole thing would be even funnier if turned out that Saunders is actually a werewolf.
---------------------------------------------
And lo, the ground breaking roll-out of the massive shake-up etc (TM New Labour) of the Coalition's flagship policy started today, which applies initially only to non-disabled childless single people living in Ashton-under-Lyme who start claiming unemployment benefit in future. I admit that BobE and I invented the bit about it only applying to people who like listening to King Crimson, flower arranging and home baking.

IDS keeps saying that the glorious new system "will mean people are always better off in work than on benefits" which might or might not be true, and it is probably slightly less untrue than under the current welfare system, which will continue for claimants everywhere else in the country (or to existing claimants in Ashton-under-Lyme who are either in a couple, have children or are disabled) for the next few years.

But how much better off will UC claimants be if they manage to find a job? Will they have a marginal withdrawal rate of 65%, 76%, 79% or 85%? In other words, will they keep 35p for every £1 they earn, 24p, 21p or 15p? A couple of clues here.

Place your bets here or use the widget in the sidebar.

Means testing and the Laffer curve

A point I have made many times: if we accept that e.g. 50% income tax is the very top of the Laffer curve in terms of maximising revenues (which should not be a policy goal anyway, but that aside), then the same applies to lower earners.

If we see means-testing as a way of reducing the cash cost of the welfare (which shouldn't be a policy goal either, long story) then the cost minimising rate of benefit withdrawal + PAYE on lower earners should also be restricted to 50% (which can be easily achieved by issuing claimants with K-codes).

BobE has spotted somebody coming at this from a different angle...

50. Planeshift
"@22 No doubt 18% was the top of the Laffer curve too."

Evidently Iain Duncan Smith thinks 65% is the top of the Laffer curve – because this is the taper rate for universal credit – which he is spinning as being about creating inventives to work.

(incidently, the DWP's impact assesment actually shows that whilst 1.2 million will have a greater incentive to work under universal credit, 2.1 million people will have less of an incentive to work due to this taper rate being higher. Meaning the spin is complete bollocks – and needs to be challenged)

How about this for a policy?

The top rate of tax should be set at exactly the same level as the taper rate for universal credit. Lets see how long the tories support a 65% rate then!

From: Thatcher's key legacy – 'trickle down economics' – may come to an end sooner than later

Reader's Letter Of The Day

From yesterday's the FT:

Sir, Helicopter money can be disbursed most quickly by cutting value added tax rates, especially in the pre-Christmas period. The greatest multiplier effect within the UK arises from reducing VAT on the sales of services that are labour intensive such as hospitality, repairs and bespoke goods.

Andrew Dundas, Perth.

I'm not sure if it's correct to talk about the 'multiplier effect' (to the extent such a thing even exists) of a tax cut, surely what he means is that VAT pushes businesses close to or past the top of the Laffer Curve. Commonsense tells us that businesses whose gross profits or 'value added' are liable to an extra 16.66% tax before they even think about paying the 40% PAYE or 24% corporation tax are most likely to be in this position.

So assuming constant business income and that all profits are paid out as wages, scrapping VAT would bring down the overall marginal rate down from 50.3% to 40.2%, which would make a huge difference to the business/employees and only a very modest overall fall in tax revenues - the receipts from a 40% are not 40/50 of those of a 50% tax, they would be 43/50 (or whatever).

[Cue idiots shouting: "But the consumer pays the VAT!". Yes, that's what the One Per Cent want you to believe - and even if it were true, why is this A Good Thing?]

To back this up, HMRC have published some handy tables on the tax gap, which shows that in relative terms, the most-evaded tax is duty on hand rolling tobacco (38%) but in absolute terms, the biggest shortfall is with VAT (they collect about ten per cent less than they expect every year = £10 billion a year out of £100 billion).

And of course the lowest tax gaps are for land-related taxes, Business Rates has collection costs plus losses of 1.5%.

The SDLT-gap is given as £0.5 billion, which is about a tenth of total revenues. The collection costs of the old flat 1% rate were famously minimal, because few people bothered evading it, but I'm quite sure that the collection costs of the savage 3% - 15% rates are much higher - it is widely observed that transactions tend to bunch at prices just below the threshold - and from professional experience I'd guess that the 'compliance costs' (i.e. 'avoidance costs') are absolutely enormous. For example, fancy firms of lawyers offer SDLT-saving schemes, for which they charge you a third of the SDLT saved.

"HM Revenue & Customs: The Exchequer effect of the 50 per cent additional rate of income tax"

I've reverse engineered the calculations they made in explaining why cutting the 50p top income tax rate to 45p would probably be revenue-neutral (see their calculations and explanations on page 15 and the chart on page 51) and I've embedded the key part of the spreadsheet below for your entertainment.

Their aim seems to be to set the income tax rate at the revenue-maximising rate, there are a few of problems with this:

1. They accept 12.8% Employer's NIC and 1% Employee's NIC as a given, even though these have both increased by 1% and are just another layer of income tax.

2. They cheerfully admit that they have no idea what the elasticity of supply of labour is (which they refine slightly and refer to as 'Taxable Income Elasticity' or TIE). On pages 18 and 19 they summarise lots of studies from various countries which give figures between 0.2 (people continue working and paying tax, even if wages fall and/or tax rates increase) and 0.8 (people will work less, avoid or evade taxes, move abroad etc if wages fall and/or tax rates increase), ignoring outliers. They can't make up their minds whether to use 0.35, 0.45 or 0.55 (so their chart uses all three).

3. As the calculations show, this is the most important unknown of all; if you assume a low sensitivity of 0.2, then the revenue-maximising income tax rate (i.e. the top of the Laffer curve) is 67%, and if you assume a high sensitivity of 0.8 it's 35%. Of course, different people respond differently; the rate might be different for high earners and low earners; the short term responsiveness might be more or less than long term etc.

4. They say that TIE can be split up into three things: labour supply (people work less or move abroad), tax planning (people work just as much but use legal loopholes) and evasion (people work just as much but do it cash-in-hand), and say the labour supply effect is "between one-third and one-half of the total TIE". This is their roundabout way of admitting that taxes on income have dead weight costs.

5. Those dead weight costs are also an incredibly important consideration. My spreadsheet shows the fall in potential GDP in the last row, i.e. the fall in declared taxable income multiplied by 0.4, which seems to come out at between 10% and 15%. That's a heck of a lot of lost output and unnecessary unemployment, by the way.

The figure for total revenues assumes that GDP starts at 100, with a flat income tax rate of 10% and National Insurance still in place (even these low rates must depress GDP by five per cent or so, let's gloss over that).

6. Another thing they ought to consider is that the cost of welfare is just an upside down Laffer curve; the way to minimise the cost of means-tested benefits is to find out what the revenue-maximising tax rate on lower earners with a small independent income would be and then to use that as the marginal withdrawal rate.

As a matter of fact, marginal withdrawal rates (especially for people who have tax deducted and benefits withdrawn at the same time) are between 70% and 100%, which is way beyond the top of the Laffer curve - in other words, we could reduce the cost of means-tested welfare by reducing income-based means-testing, although that point seems to escape them.

Laffer Curve Fun, with a side serving of nepotism

HM Revenue & Customs has finally come up with a sixty-page report explaining why total revenues will not fall if the 50% income tax rate is reduced to 45%, here's the chart showing possible Laffer Curves from page 51:(image pinched from The Guardian, that's a good article if you ignore the leftie spin).

Fair enough, but what is this TIE of which they speak? It's short for 'Taxable income elasticity', see page 14:

The measure of the overall responsiveness of total taxable incomes to changes in marginal tax rates is the "taxable income elasticity" (TIE). This estimates the percentage change in total taxable incomes in response to a one per cent change in the net-of-tax rate (the proportion of each additional pound earned received by the individual after tax, also known as the marginal retention rate), and therefore captures all the behavioural responses described in Chapter 2.

Chapter 2 isn't much use either, but basically TIE is some sort of estimate of the price-elasticity of supply of labour. So if TIE is low at 0.35, people don't drop out the (taxpaying) workforce in response to lower post-tax incomes, and the increase in tax revenues by nudging the rate up to 50% outweighs the effect of the smaller tax base. If TIE is high, at 0.55, then once you get to an income tax rate of 43%, so many people drop out of the (taxpaying) workforce that total revenues start to fall if you nudge the rate any higher.

But HMRC cheerfully admit that they have no idea what the average TIE of higher earners is, so the whole thing is fairly useless, except to illustrate the point that a reasonable man concludes that while there must be a revenue-maximising income tax rate:

a) Whatever the true TIE is, the revenue-maximising tax rate could be anywhere within a very wide range (between 43% and 57%, going by the chart) and caution dictates we ought to go for the lowest figure in that range (43%).

b) We do not want to be anywhere near the revenue-maximising income tax rate anyway, we would rather be at the economy-maximising income tax rate, which happens to be zero per cent.

For your side-serving of nepotism, see City AM.

Reduced VAT on pubs and restaurants

From today's Evening Standard:

More than two dozen of Britain's top pub and restaurant chains have joined forces to campaign for a cut in VAT to boost the country's ailing leisure industry. (1) They have joined together to back French hospitality entrepreneur and lobbyist Jacques Borel in his campaign to get VAT reduced from 20% to 5% on food, drink and accommodation in the UK.

Borel successfully battled for the cut in VAT in France that not only led to thousands more jobs in the leisure industry but also saw the government's tax take from the sector actually rise as more people went out... (2) Belgium, Germany and Sweden have also cut VAT in the leisure sector. Ireland joined the group last summer, cutting the tax from 13.5% to 9%. (3)

1) Well, that gives lie to the myth that "the consumer pays the VAT and it doesn't affect the supplier" doesn't it? If it made no difference to the supplier, then they wouldn't care, would they? If you look at actual facts rather than listening to politicians, you'd know that about two-thirds of VAT is borne by the supplier, and with discretionary spending like pubs and restaurants, it might be nearly all of it. And so that actually puts businesses out of business, unlike corporation tax which can't possibly push a low-profit business into making losses.

2) Yup, we also observed this with the car scrappage scheme, that was to all intents and purposes a refund to manufacturers of the VAT they would otherwise have to pay. The extra VAT receipts were slightly more than the cost of the scheme (so why didn't they just exempt new cars from VAT, you might ask...), and you can add the extra corporation tax, PAYE and dole money saved on top of that.

It's because having an extra 16.66% tax on gross profits ('value added tax') is sufficient to push the marginal rate past the top of the Laffer Curve. Part of the reason for the increase in tax receipts from one or the other industry which benefits is because consumers change their spending patterns in response to the slightly lower price and much greater output, so it would be unduly optimistic to assume that if VAT were abolished entirely that ordinary tax receipts on profits and wages (and dole money saved) would make up the shortfall, but even the most pessimistic assumptions (see my post on Laffer Rainbow) say that scrapping VAT entirely (revenues £100 billion per annum in the UK) would only lead to a £50 billion fall in total tax revenues. So the chances are that the growth in the overall economy would be roughly a much as the VAT cut and a smaller share would be channelled via the government. What's not to like?

3) So we'll have plenty of raw data for doing before-and-after comparisons.

Tobacco duty and The Laffer Curve

There was an interesting debate in the comments at VGIF. Here's my adaptation of Pub Curmudgeon's theory on how countries set their tobacco duty rates...

It is clearly impossible for every country to set the duty rate a little bit higher than their neighbours, neither is it possible for every country to set it lower than their neighbours.

Assuming equal size countries, the rate would be the same everywhere (with a lower upper limit for booze than for tobacco). We can then introduce your perfectly valid observation about the ideal rate being lower in small countries and higher in large countries and we lead to some sort of equilibrium.

My caveat would be that it's not 'size of country' that matters as much as 'how easy it is for your citizen's to go abroad' i.e. UK and France are similar size/population, but we Brits are stuck with buying in the UK, the French can go to E, CH, D or Benelux quite easily.
----------------------------
So let's see whether this stacks up in real life...

The price of a packet of twenty in various European countries is relatively easy to find, e.g. here, plus two missing ones from here. Populations are easy to find e.g. here.

The resulting scatter graph looks like this:The generalisation that big country = high tobacco duty clearly holds for the seven largest countries. In order from left to right these are Romania, Poland, Spain, Italy, UK, France, and Germany, the coefficient of correlation between population and price per packet is quite high at 0.76 (see footnote).

But for the smaller countries with populations under 20 million, there is no correlation whatsoever, and the coefficient of correlation for the whole data set is only 0.13, practically meaningless.
----------------------------
So let's strip out the 'big seven' and do another chart plotting price per packet against 'how easy it is for a country's citizens to pop over the border to buy cheaper elsewhere' for the remaining countries. The simplest way of measuring 'how easy it is [etc]' is to count the number of land borders a country has with other countries. The resulting chart looks a lot more promising - rather unsurprisingly, the most expensive cigarettes are to be found in countries which have no or only one land border(s) with another country (Iceland, Malta, Ireland, Norway).

The coefficient of correlation for this sub-set is -0.55, not staggeringly high but enough to mean something:Footnote: The 'border' effect also conveniently explains why line for the big seven diverges at the top. Germany is larger than the UK but it has land borders with nine other countries whereas the UK has none (ignoring the border between Northern Ireland and Ireland). The 'border' effect thus has a stronger impact than the size effect and cigarettes are cheaper in Germany than in the UK.
-----------------------------
Finally, we can put the two theories together.

Let's assume that the price per packet is EUR 9, reduced by EUR 1 for each border a country has, plus EUR 1 for every 15 million in the population of that country, and then plot price per packet against the result. The constants 9 and 15 million are arrived at by observation, trial and error* so as to end up with the highest possible coefficient of correlation, which is 0.52. The resulting chart looks like this:* For example, Germany has borders with nine other countries, so the border element of its tobacco duty is EUR 9 minus 9 x EUR 1 = EUR 0. It has a population of 82.2 million, divided by 15 million = a population element of EUR 5.48, EUR 0 + EUR 5.48 = EUR 5.48. The actual price of a packet of cigarettes = EUR 5.16.

At the other extreme, Malta has no land borders, so has a full border element of EUR 9. It has a population of 400,000, divided by 15 million = population element EUR 0.03. EUR 9 + EUR 0.03 = EUR 9.03. The actual price of a packet of cigarettes = EUR 9.

Hungary is somewhere in the middle. It has borders with 7 countries, so the border element is EUR 9 minus 7 x EUR 1 = EUR 2. It has a population of 10 million, divided by 15 million = EUR 0.67. EUR 2 + EUR 0.67 = EUR 2.67. The actual price of a packet of cigarettes = EUR 2.51.

Killer Arguments Against LVT, Not (148)

Just to show that the idiocy goes right to the top, here are two paragraphs from HM Treasury's "Tax policy making: a new approach" as reported in Hansard:

11. In this inquiry, we received many submissions advocating radical change to the tax system, such as the imposition of a land value tax. (1) The supporters of such a tax consider that it would tax economic rent rather than economic activity and would meet the OECD criterion that recurrent taxes on immovable property were the least harmful tax. However, as the CBI notes, "the OECD acknowledges that it is politically difficult (2) for governments to shift the tax base onto property." (3) The ICAEW warned "Our initial conclusion is that, even if such a move was desirable economically (4) and let alone whether it would be politically acceptable,(5) it would involve a major rebalancing of the UK tax system which would take time to achieve (6) and risks introducing considerable distortions and behavioural changes." (7)

12. Not only are there political difficulties: practical matters, such as the way in which such values would be assessed and the extent to which such a tax should take account of the current or the potential use of land, would also need careful consideration. (8) We also note concerns that "While such a tax system would avoid distortions in economic behaviour, (9) it would be highly unlikely to yield sufficient revenues to fund socially useful expenditure (10) without producing substantial inequity." (11)

1) Oops. I think I missed that one.

2) It is only "politically difficult" because the government and the powers that be generally have spent decades brainwashing people into thinking a) that taxing economic activity and wealth creation is a reasonable way of doing things and b) that house price rises are good for us.

3) I wish they'd say "land and buildings" to make it clear what they are talking about. In any event, the CBI ought to be speaking from the point of "British Industry" who are perfectly accustomed to paying Business Rates, which are so close to Land Value Tax as makes no difference.

4) It's "were" not "was" and actually it "is" and they must know that.

5) See (2).

6) Everything takes time, it all depends which taxes you replace first. So let's start by rolling all existing taxes which relate to residential land and buildings or 'wealth' generally into a flat tax on residential land values. The only real constraint on how quickly we did this is how quickly people can be de-brainwashed.

7) No! It's the current system which creates "considerable distortions", the fact that people would behave differently if there were no taxes on economic activity and only taxes on economic rent is an argument in favour of the latter.

8) They already have two models - Business Rates for commercial land and buildings and Domestic Rates in Northern Ireland. All the info we need is already held by HM Land Registry or on the Council Tax register after that it's just a question of bunging in the [current tax rate + a percentage of current selling prices] in each defined area to get a fair approximation of the rental value of land in each area, you tot these up to give you the rental value of all residential land in the UK (the tax base, Y); you then decide a figure for how much tax you want to raise (X), divide X by Y to give you the tax rate, apply X/Y to [the local rate x size of each plot in each area] and we're away.

It'll never be scientifically perfect, but any over- or under-estimates will iron themselves out, but so what anyway? Does the rate of VAT automatically adjust itself down so that marginal businesses are kept afloat, does it automatically adjust itself up on businesses which appear to be making super-profits (banks, from 2000 to 2007 or thereabouts)? I think not. Would it not be better to have a tax which allows 'the markets' to decide the rates in the same way as we currently decide rents and prices?

9) That's not what they said at (4), is it?

10) Wot? Do they not realise the circularity involved?

a) Even under current rules, the residual rental value of UK land is about £150 billion per annum. If we taxed that at 100% instead of just collecting £50 billion in Council Tax and Business Rates we could get rid of VAT and merge income tax and National Insurance into a flat income tax of about 30% with a generous personal allowance. Would that not be "socially useful"?

b) Shifting from taxes on activity to taxes on rents gets rid of the dead weight costs caused by the former, and this extra growth goes £ for £ back into higher rental values, so it's a virtuous circle.

c) Forecast total tax revenues for 2011-12 are £531 billion (excluding booze, fags and fuel duty), that works out at an average £20,000 per household (including taxes borne by 'businesses' which are indirectly borne by households). If the average household no longer has to pay £20,000 in income tax, VAT, NIC etc, there is no reason to assume that they wouldn't be able to pay that much in LVT.

d) Then there is the invisible half of the Laffer Curve which few people talk about. Although income tax revenues would be zero if the income tax rate were zero per cent, by how much would the economy grow? A tenth? A fifth? A quarter? There's only one way to find out! Most of that growth would go into higher land rental values, so in effect, the land value tax would pay for itself, and a household's average net income would increase by something approaching £20,000 per annum.

11) Even if it's only half that, well it's not to be sniffed at, is it? Deliberately depriving every household of £10,000 or £20,000 income every year out of political cowardice seems like "substantial inequity" to me.

Treasury Tomfoolery

There's been an outbreak of common sense at the Federation of Small Businesses. From the BBC:

The FSB is urging VAT be cut to 5% in the construction and tourism sectors. "Consumer demand is a key barrier to economic growth so such a cut would encourage people to spend in these areas," the FSB said in its Voices of Small Business Report.

The FSB said: "Evidence from other EU countries shows that any lost revenue to the Exchequer by making VAT cuts will be met by earnings from additional demand, jobs and the wider economic activity."

Although the basic EU rule is that the standard rate of VAT has to be at least 15%, it appears that countries can reduce the rate on specific sectors, which is what Ireland seems to have done recently. So far so good. The depressing bit is right at the end of the article:

A Treasury spokesperson said: "Reduced VAT rates of the kind suggested would make a significant impact on revenue. Any claim that a boost to foreign tourism or construction would outweigh these effects would need to be looked at very carefully indeed."

In other words, they didn't give these secondary effects any thought whatsoever when they hiked VAT from 15% to 17.5% to 20%, did they? Even Ed Balls seems to have finally grasped that you cannot keep merrily increasing VAT and expecting overall tax receipts to keep going up, for crying out loud.

Fuel Duty Fun

From The Daily Mail:

The Treasury is facing a £637 million deficit after fuel sales dropped by one billion litres this year, the AA revealed.

Service stations in the UK sold 835 million fewer litres of petrol and 247 million fewer litres of diesel in January to March 2011 compared to the same period three years earlier. The [15.2] per cent dip in petrol sales and the 6 per cent fall in diesel sales were caused by higher fuel costs and consumers tightening their belts.

Shock horrors! Fuel prices up and sales down! The demand for fuel is price-elastic! Moreover, the demand for diesel (primarily lorries, so we'd expect diesel sales to fall in line with sales of food and so on) is far less price-elastic than the demand for petrol (which is discretionary spending to some extent). Well who'd have thought?

More to the point, the £637 million tax shortfall is mathematically correct but logically flawed. Let's stick those figures in a spreadsheet:

First quarter 2008:
Million litres of petrol sold = 5,493
Million litres of diesel sold = 4,117
Total = 9,610
Average price (from here) = 106 pence/litre
Fuel duty = 53.65 pence/litre

Change:
Petrol sales down 15.2% = 835 million litres
Diesel sales down 6% = 247 million litres
Total = 1,082

First quarter 2011:
Million litres of petrol sold = 4,658
Million litres of diesel sold = 3,870
Total = 8,528
Average price (estimate) = 135 pence/litre
Fuel duty = 58.95 pence/litre

So how much is the fall in tax revenues?

If we just multiply 1,082 million litres by 58.95 pence, we arrive at £638 million (the AA's figure).

Oh no, it isn't!

The £637 million figure is of course complete nonsense and is comparing apples with pears on several levels - for example, they overlook VAT - which used to be 17.5/117.5 of a smaller number and is now 20/120 of a bigger number.

So how much is the increase in tax revenues?

The true tax per litre in the first quarter of 2008 was 69 pence (15p VAT and 54p fuel duty); in the first quarter of 2011 it was 81 pence (22p VAT and 59p fuel duty). So the AA could argue (if it so wished) that the tax shortfall is £881 million (1,082 million litres x 81p), but it would be far more correct to say that total revenues were £273 million higher in first quarter 2011 than in first quarter 2008:

Q1 2011: 8,528 million litres x 81p = £6,946 million
Q1 2008: 9,610 million litres x 69p = £6,673 million
£6,946 million - £6,673 = £273 million.

So it's quite clear we are still not past the top of the Laffer Curve (although we may be getting near it) and there is no 'revenue shortfall'. Whether you think that fuel duty is a better or a worse tax than other taxes is a separate issue; to my mind it's like Land Value Tax for roads, so is on the "good" side of the line, as opposed to VAT generally, income tax, National Insurance etc, which are on the "bad" side of the line.

Just sayin', is all.

Laffer Fun

As well as putting his name to the Laffer Curve, another of his great observations was that tax receipts fall by far less than you'd expect when tax rate cuts are announed in advance (and rise much less than you'd expect when people know that rates are going to go up) because of timing differences. This is largely a one-off thing and not to be confused with the general Laffer effect (i.e. the increase in economic activity wholly or partly balances out cuts in tax rates and vice versa).

There was a fine example in today's City AM:

BT HAS paid the latest instalment to its pension deficit programme nine months early in order to qualify for the tax deductible corporation tax rate of 28 per cent, rather than the 26 per cent rate that will apply from April.

BT says the decision to pay £505m into the BT Pension Scheme, the actuarial value of the £525m due to have been paid in December, made financial sense. The timing of the tax deduction will also be brought forward to the first-half of the upcoming financial year.

Tax rises & behavioural change

A post not from the desk of Mark Wadsworth:

Reports are out today of a January surplus £700 million higher than forecast: Daily Mail, Guardian

What they gloss over though is the main reason for income tax receipts being so high.

Alastair Darling & Gordon Brown being the intelligent sort of chaps they are introduced a 50%(42.5% on dividend) top rate of tax for incomes over £150,000.

As people were sufficiently forewarned of this, those who own & operate private limited companies and pretty much have total control over their reported income in any given year took massive dividends in the 2009/10 tax year. They did this in order to pay the tax at 32.5% in January 2011 (hence the surplus) rather than pay it at 42.5% in January 2012.

The Laffer Curve, GDP and the rental value of land

Here's the Power Point presentation I've been working on for the past couple of weeks:

The real enemy is the civil service

I printed off the DWP publication on their proposals for a Universal Credit, and a lot of it is to be welcomed, such as this on page 8:

Currently, when combined with tax and National Insurance payments, the withdrawal of Tax Credits, Housing Benefit and Council Tax Benefit can lead to Marginal Deduction Rates which are nearly 96 per cent, much higher than the highest rate of Income Tax... The combined effect of benefit withdrawal rates and additional tax as earnings increase is called the Marginal Deduction Rate and has the same practical effect as a tax rate.

Correct! It is to all intents and purposes taxation.

1. Let's imagine the Kind Tooth Fairy (the DWP) insists of giving every citizen a bit of money so that they don't starve, and has no method of clawing any of it back again (a bit like Child Benefit or the basic state pension); and the Wicked Tooth Fairy (HM Revenue & Customs) has to collect as much money as it can by deducting it from people's wages.

2. If the Wicked Tooth Fairy knows about the Laffer Curve, he would know that the revenue maximising rate on lower earners is about 60%, which he could arbitrarily split up into 12.8% from the Employer and 47.2% from the employee, call it 50% from the employee for sake of argument, which the Wicked Tooth Fairy would then ask employers to deduct from people's wages and send straight to him.

3. So the cash-cost minimising MDR is, by and large, the same as the revenue-maximising tax rate on the Laffer Curve, which sure as heck ain't 80% or so*.

4. The civil servants have deliberately spiked the whole point of the scheme by having a MDR of 65% or 76% (which I looked at yesterday), but, despite having admitted on page 8 that means testing is to all intents and purposes the same as taxation (in other words could be deducted straight from people's wages, to the extent they have any), they write this in the Executive Summary:

For those in employment, Universal Credit will be calculated and delivered electronically, automatically adjusting credit payments according to monthly income reported through an upgraded version of the Pay As You Earn tax system (on which HM Revenue & Customs will be consulting shortly).

The system will be simpler and will respond more quickly to changes in earnings so that people will not face the same complexities as they do now, particularly at the end of a tax year. As a result people will be much clearer about their entitlements and the beneficial effects of increasing their earnings by taking on more hours or doing some overtime...

This would involve an IT development of moderate scale, which the Department for Work and Pensions and its suppliers are confident of handling within budget and timescale.

AAAAARGH!!

5. How is that in any way better, or cheaper or less prone to fraud and error than simply giving welfare claimants £64 a week (or whatever the amount will be) and a PAYE code with no personal allowance, which tells employers to deduct 50% income tax and no back chat? The regular payments don't need to 'respond to changes in earnings', the claimant knows that he has £64 a week coming in (or whatever) and knows that for every £1 he earns legitimately, he will keep half.

6. And do you know what? We already have PAYE codes like this, they are called K-codes, it's all routine stuff that any payroll lady can handle (in fact, it's the easiest PAYE code of all to deal with, even easier than a BR 'emergency code' as the 50% covers both income tax and Employee's National Insurance.)

* Once Employer's NIC has gone up to 13.8% and Employee's NIC to 12%, for every £113.80 the employer pays out, a worker on the Universal Credit would keep £23.80, i.e. £90 makes its way back to the government, which is a tax rate of 79%. It's even higher if you add the VAT on top of the wages.

Those Welfare Reforms: Epic Fail (1)

Some of the blurb in the full 73 page document is most promising, because rolling most existing welfare payments into the the Universal Credit/Single Universal Taper scheme is a sizeable step towards a 'Citizen's Income'. For example, from page 55:

Establishing a single withdrawal rate, and eliminating the hours rules currently present in Working Tax Credit, has the potential to create a much more flexible labour market. Workers will be able to work the number of hours that most suits their needs and those of their employer, without being constrained by the structure of the benefits system. Employers will find that their workforces become more flexible and open to opportunities for progression.

For example, it will now be financially rewarding for a lone parent to work 15 hours per week, or 17 hours per week (both of which would not have been financially rewarded under the existing system which only recognised 16 hours per week);
and should more hours be available, the extra earnings will no longer face a Marginal Deduction Rate [MDR] of 96 per cent....

OK, they're focusing on lone parents. The MDR under the UC/SUT is explained on the previous page:

We expect the impacts on earnings incentives to be large. Universal Credit will improve earnings incentives for 700,000 current low earning workers. It increases their financial reward to work more hours or increase their wage rate, and as such
it creates a more flexible workforce.

The expected Universal Credit withdrawal rate of 65 per cent means that to all intents and purposes, the highest Marginal Deduction Rate 24 for low-earning workers would be reduced from around 96 per cent to 65 per cent for those earning below the personal tax threshold and to around 76 per cent for basic rate taxpayers*.

OK, 76% doesn't sound quite as bad as 96%, but that's a very misleading comparison. If we now flip to the last set of Tax Benefit Model Tables, Table 1.3a (Lone Parent with 2 children under 11, no child care costs) we see that the MDR is between 85% and 90% on earnings up to £180 a week, so it'd be a vast improvement for these lower earners BUT the MDR is then a flat 70%** all the way up to earnings of £520 a week (or £27,000 a year, far more than the median income - this figure would be even higher if the table included the 30-hour element).

I don't dispute that overall the UC/SUT will be an improvement, but as ever they are taking from the 'better off poor' to give to people further down, and anybody who thinks that 65% or 76% is the cost-minimising MDR needs his head examining (means testing is just like taxation, so the tax revenue maximising tax rate from the Laffer Curve must be the same as the MDR which minimises the cash cost of welfare payments).

* If you earn £100 and have £31 PAYE deducted, your net pay is £69, and they take away 65% of that in means testing = £45, so you end up with the princely sum of £24 extra in your pocket.

** UPDATE: SW in the comments reminds us that the MDR on basic rate taxpayers claiming Tax Credits is going to go up from 70% to 73%.

My email to the DWP

Dear Mr Spiers

I note from an article in the Daily Mail on the Universal Credit/Single Unified Taper that "A withdrawal rate of between 60 and 65 per cent is being debated, according to sources."

Obviously, you are not allowed to tell me whether or not this is correct etc, but allow to me to point out (as I possibly did not emphasise enough in my original submission of early August)...

1. Means testing of benefits is exactly like taxing income, so we have to think about the Laffer Curve.

2. I believe that the revenue maximising tax rate on lower/median earners is about 60%.

3. Therefore the cash-cost minimising benefit withdrawal rate must also be about 60% of income.

4. Unless people earn very little indeed, their employment income is liable to Employer's NIC at a marginal 12.8% (or 13.8% next year) which we can round to 10% for the sake of this discussion. This is of course legally borne by the employer but economically it is borne by the employee - a 1% increase in the rate leads to a 1% fall in headline wages (all things being equal).

5. Therefore, the cash-cost minimising withdrawal rate after deducting the 10% Employer's NIC must be about 50%.

6. 50% is a very convenient figure, as the PAYE system is already geared up to deducting tax/Employee's NIC at a flat rate of 50% of cash salary (ignoring bands, personal allowances) under the K-code system.

7. So we could achieve a cash-cost minimising withdrawal rate of 60% overall AND obviate the need to have a separate system of means testing by paying people their Universal Credit and giving all adults in the claimant household a K-code for PAYE purposes - the benefit withdrawal would be dealt with via the PAYE system, more or less for free as far as the taxpayer is concerned and at minimal extra hassle for the payroll department.

I look forward to your response and remain


Mark Wadsworth
------------------
UPDATE: from today's Evening Standard:

However, according to the Times, Mr Duncan Smith has now persuaded the Treasury to let him claim up front “a large chunk” of some £9 billion of potential savings (1). His new system would carry a guarantee that anyone taking a job would be better off than if they were on the dole, through being allowed to keep more of their benefits.

But over the summer it emerged that the senior Treasury official responsible for checking his calculations, Claire Lombardelli, did not believe his sums were credible. (2) The Chancellor has made clear the cost of benefits must fall, even if it means putting off major reforms on the grounds that they would be expensive to introduce (3). Cutting welfare is vital to his plan to cushion other departments from cuts. (4)

The Department for Work and Pensions said: “We are working closely with the Treasury... We are all agreed on the urgent need to reform the welfare system and help more people into work and off benefits.”(5)

1) Correct. The DWP have running costs, i.e. civil servants' salaries of £9 billion. That's our next port of call.

2) Well she would say that, wouldn't she? See (1).

3) That's a pretty twattish thing for a Chancellor to say: "I'm sorry darling, let's put off that roof repair that will cost a few hundred quid, even if it means that we'll get a bill of thousands of pounds later on, once all the woodwork is nice and rotten."

4) An inevitable knock on effect of reducing the marginal withdrawal rate is that more people will be working, so they'll be claiming less and/or paying more tax, so the overall cash cost goes down. At the same time, the incomes of low earners will go up. Win win. What's not to like? As to "cushioning other departments from cuts", I'm not sure that there'll be any departments left once my Bloggers Cabinet is finished.

5) Well they would say that, wouldn't they? See (1) and (2). Let's see how good they are at helping themselves into work once the whole welfare system runs itself.

Mr Corporatist Shill speaks

Allister Heath in CityAM:

IMF is right... The IMF is upbeat for 2011, predicting growth of two per cent. It makes it clear that slashing the deficit and controlling public spending is the only way forward to avoid a sovereign debt crisis. While squeezing state spending will slow growth (1), the IMF is confident that liberating resources for the private sector (2) will ensure that the recovery continues strongly.

More emphasis should be given to reducing public sector compensation wage premia (3) and achieving savings in benefits through better targeting (4), it argues, policies that will be music to Osborne’s ears as he prepares for next week’s party conference in Birmingham.

1) No it won't.

2) Let's never forget that in 2009-10, public sector salaries and pensions were £169 billion; the total cash cost of welfare and old age pensions was £217 billion and subsidies and other payments to supposedly 'private sector' organisations were £281 billion. If we are serious about reducing government spending, then this is the first place to look - cut all this by half, and hey presto, the annual deficit of £150 billion-odd would be more or less eliminated.

There's no need to worry about strikes or anything, because those private organisations that are providing useful goods and services would not be affected and their workers are largely non-unionised. What's not to like?

3) Sure, only a third of public sector workers are 'front line'. So we can sack a couple of million with no detrimental impact on 'front line services', which are mainly provided by lower paid public sector workers anyway. Thus this automatically reduces average public sector pay. So let's get on with it.

4) The easiest way to reduce cash welfare payments is to keep basic entitlements much the same (OK, cut them a bit for 'single mothers') and reduce marginal withdrawal rates from their current 70% to 100% range down to something humane (no more than 50%). The Laffer Curve tells us that the cash cost of welfare will fall even though the incomes of those in low paid jobs will rise. What's not to like?

And the IMF are a bunch of shysters anyway, sod them and what they say.