Thursday, 19 February 2015

Killer Arguments Against LVT, Not (354)

One genuine problem is that we LVTers argue and bicker between ourselves too much over relatively minor matters (while agreeing on the broader principles). So the Homeys can play divide and conquer. I pointed out that in the grander scheme of things and at current price and rent levels, the site premium of most homes is in the range of 3% - 3.5% of its selling price.

Physiocrat: Why is it even necessary to think about selling prices at all? There is so much rental evidence on the property websites.

One of the things you will notice is that there is a "floor" to these rentals - for example, you will not find a three-bed semi to rent anywhere for less than around £500 pcm. As a first approximation, it could be assumed that the land value in these instances, which covers a substantial area of the scabbier parts of the UK, is zero.

This assumption also does away with the argument for homestead allowances and other complications - the allowance is built into the valuation on the basis that the land value is too small to measure or be worth collecting.


In principle he is quite correct and I do not disagree (apart from the word 'scabbiest'), but let's look at the numbers:

The purist approach

A quick search on Zoopla tells us that you the rent for a three-bed semi in one of the cheapest areas (TS6) is indeed about £6,000 a year gross (£450 a month plus council tax).

Knock off £4,000 a year actual costs (amortisation of bricks and mortar, repairs, insurance etc) and the site premium here is about £2,000.

The rough and ready approach

Three-bed semis in TS6 sell for as little as £70,000 - £75,000.

£70,000 x 3.5% = £2,450 site premium.

I rest my case.
----------------------------------
Clearly, there must be a few little houses in the most run-down streets in the areas with the lowest wages and highest unemployment where the site premium is zero or close to zero, but that's only a few thousand houses out of millions and not worth worrying about. Such houses sell for £20,000 or something, and 3.5% of not much is next to nothing. £700 a year is the marginal cost of pavement repairs, refuse collection and street lighting if nothing else.

The other point is that when somebody buys an existing home, what he is paying for is future profits i.e. gross rental value minus running costs i.e. the 'site premium'.

Therefore homes trade for a multiple of their site premium and that multiple is surprisingly similar for all homes except the bottom one or two percentiles and the top one or two per centiles. But who cares? You can't base a sensible tax system on outliers; you start in the middle and work outwards.

14 comments:

Ben Jamin' said...

I think Phys may believe that you think land rents are 3.5% in perpetuity.

Whereas, you've used it as a simple way of calculating them as today's current selling prices.

If you fixed 3.5% into bands today, land rent wouldn't be 3.5% of selling prices tomorrow.

But, as long as that 3.5% is fixed into bands, it doesn't matter.



Bayard said...

"The other point is that when somebody buys an existing home, what he is paying for is future profits i.e. gross rental value minus running costs i.e. the 'site premium'"

But in that case, if the site premium is zero, then the house should be free. Even though the brix 'n' mortar depreciate, and cost a lot to maintain, they are still worth something, although I would agree that there are cases where the running costs on a house are so high that the B'n'M element can have a negative value. One hears for instance of listed buildings in poor condition being purchased for £1.

Physiocrat said...

The Vale of White Horse study got into a mess due to the use of selling prices. The selling prices of property subject to UBR is depressed relative to that subject to CT. The value that needs to be established is Gross Annual Value, that is rental value plus tax already paid. To adjust capital values to achieve the correct relativities would be necessary, ie the capitalisation of existing property taxes would have be added on.

A useful desktop exercise would be to look at rents within parts of the NE postcode area: Newcastle upon Tyne, North Shields. This comprises the city centre and pockets of good residential areas in Jesmond (NE2), Gosforth (NE3), and parts of Tynemouth (NE) and Whitley Bay (NE25). In most of the rest, land values are bumping along the bottom.

It would be interesting to map location premium values by postcode eg NE2 X--.

Lola said...

Isn't it interesting how a lot of these 'rules of thumb' actually work out 'about right'? Which is what you'd expect, the rules of thumb being the learned and aggregated wisdom of the whole crowd.
I am very much a 'keep it simple' bloke myself.
3.05 to 3.5% site premium works for me.
Just please let's get on with it, eh?

Kj said...

What Phys says: the exercise of a percentage of current property values may be a good approximation now, but after a significant LVT has been implemented, it'd be much more complicated. And the Method of adding in capitalized value of taxes sounds like even more of a hassle to communicate to taxpayers and voters than rental value IMO.

Mark Wadsworth said...

BJ, L, correct, keep it as simple as possible.

Apart from that, we have engaged in an object lesson of appearing to disagree with each other, while actually agreeing 99%.

B: ".. in that case, if the site premium is zero, then the house should be free."

Correct. In a few very marginal areas, houses sell for £10,000 or something.

"Even though the brix 'n' mortar depreciate, and cost a lot to maintain, they are still worth something"

Not if they are in the wrong place. Build a house in the Antarctic and try selling it.

"I would agree that there are cases where the running costs on a house are so high that the B'n'M element can have a negative value.

One hears for instance of listed buildings in poor condition being purchased for £1."


Is that negative site premium or negative B'n'M? Does it matter?

It's not even clear if you are agreeing with me or disagreeing.

Phys, firstly, I made it clear that this rule of thumb applies to residential, not commercial premises liable to BR (a crude from of LVT).

I know all that stuff I have done all that stuff, I have used loads of different approaches with lots of sampling and interpolation and adjustments and it is a simple observable fact that - apart from the top and bottom percentile - houses sell for about thirty times their site premium.

So keep it simple when explaining to the layman. it's 3.5%, end of back chat.

Kj: "after a significant LVT has been implemented, it'd be much more complicated."

Nope, it gets easier.

Having got the ball rolling with 3.5%, you just monitor rents. If the gross rent for an average home in an area goes up £1,000 and there are three thousand homes in that area, then the tax to be apportioned between homes in that area is increased by anything up to £3 million.

The point is that the total tax to be collected is a known (if somewhat arbitrary) figure but as between all the homes, the way in which it is apportioned is fixed. See KLN number 353.

You can run parallel calculations based on prices, where the site premium = current LVT bill plus a small percentage of home value.

As a cross check, you see what typical rent for average home is and deduct £4,000 and make sure that the LVT bill is no higher than that.

It's dead easy once you've set up the spreadsheet.

Physiocrat said...

I remain to be convinced. A three storey house, 100 sq metres, in the centre of Brighton, will be around £500k. In the middle of Sunderland, you could buy a similar amount of space for less than half that.

What I cannot see is how the location value can be the same proportion of the selling price in both Sunderland and Brighton. Or have I misunderstood what is being proposed here?

Mark Wadsworth said...

Phys, you can say what your like but just go on Zoopla and look at the numbers.

3 bed terrace, BN1, sells for about £500,000. Rents for about £20,000 a year. Deduct £4,000 actual costs, site premium = £16,000.

£16,000 over £500,000 = 3.2%.

3 bed terrace, SR1, sells for about £90,000. Rents for £7,500 a year. Deduct £4,000 actual costs, site premium = £3,500.

£3,500 over £90,000 = 3.9%.

I've done a much bigger and far more scientific sample than this and 3.5% looks about right for all but the extreme cheapest and extreme most expensive.

The explanation is that what landlords are paying for is the net present value of the site premium, the pure profit element.

If a house has a gross rental value of £5,000 and actual running costs of £5,000, then why would anybody want to buy it???

Bayard said...

"If a house has a gross rental value of £5,000 and actual running costs of £5,000, then why would anybody want to buy it"

To keep their bicycles in? (not kidding, I have come across this happening). There's got to be some value in the actual structure of the house, the materials and the fact that it provides a roof over a space, even if you don't intend to live in it. So, even if the site premium is zero, it doesn't mean that the house itself is without value.

In the days before rampant land price speculation, zero site values were much more common and therefore those houses were priced much more like cars today: expensive when new, cheap when old, thrown away (or abandoned) when uneconomic to repair.

Even as late as the 60s, if you bought a farm, you paid nothing extra for the land occupied by the house over its value as agricultural land.

Mark Wadsworth said...

B: "In the days before rampant land price speculation, zero site values were much more common..."

Yes, the good old days. I know all this.

That is what I am trying to get us back to. LVT is a large part of that.

But we are still our own worst enemies, bickering over marginal situations.

I keep saying: start in the middle and work outwards, if you get the middle bit right, then top and bottom fall into place.

Physiocrat said...

Amen to the last comment. But when Labour and Conservative advocate cutting business rates to help small business there is an evident failure in understanding of basic principles, at the highest level.

Mark Wadsworth said...

Phys, but is there?

Some of them have never had it explained to them and/or really can't understand; others are allied with the FIRE sector.

Bayard said...

"an evident failure in understanding of basic principles, at the highest level."

Not at all, cutting business rates helps landowners, owner-occupiers and landlords, sterling chaps all, whilst appearing to help tenants, who are the ones who fail to understand basic principles.

Physiocrat said...

Bayard - so utter cynicism from politicians across the spectrum exploiting public ignorance.

Labour is worse because they are sailing under false colours.