Tuesday, 29 July 2014

Rebuild costs and land prices

There is one simple statistic which enables us to kill two KLN birds with one stone.

The first is that "I paid for my home out of taxed income.".

The second is a long running argument I have had with various people about the typical rebuild cost/value of an average three-bed semi as distinct from the land value.

We know from Nationwide statistics on the ratio of house prices-to-earnings, that for the period 1953 to 2001, the ratio only went above three (never quite reaching four!) twice: for a couple of years in the early 1970s bubble and for a couple of years in the late 1980s bubble.

But throughout that period, builders were happily building 200,000 - 300,000 new homes a year in all parts of the country, it is NIMBYism and land price speculation which put paid to that, we're now down closer to 100,000.

We can argue over various up or downward adjustments, but we can safely assume that builders did it to make a profit, and that their average costs per new home were less than three times an average year's salary. And we can also assume that building techniques are little changed, it's largely manual labour. The cost of copper has gone up, but the cost of white goods has gone down so it all nets off.

The average full time male annual salary is currently £26,500 (the median is much lower than that), times that by three = £80,000 (which is what I have always said).

Going back to the first KLN, yes, people paid for the bricks and mortar out of taxed income, the same as they pay for anything out of taxed income (glossing over the fact the mortgage was subsidised via MIRAS), but most people who bought before 2001 did not actually pay a penny for the land, they got that for free.

And as Land Value Tax is only on the land element, for which any pre-2001 purchaser paid effectively nothing, it is clearly not double taxation for these people.

Agreed, people who were first time buyers after 2001 will have to pay twice for the land, that's as can't be helped, but they are all still of working age and will still pay far less tax overall if we shifted taxes from earnings and output to land values.


Anonymous said...

@MW "but most people who bought before 2001 did not actually pay a penny for the land, they got that for free"

What's the significance of 2001 for this purpose?

Mark Wadsworth said...

PC, that was the last year in which the price to income ratio was three (or lower), it then rapidly went up to about five.

Lola said...

Intriguingly your 2001 date correlated nicely with my business experience. As the Brown Terror got really underway, and noting that 2001 was when the failed FSA got going, it was as clear as day that it was all going to go tits up and we stopped doing mortgage stuff, as far as possible.

Mark Wadsworth said...

L, in that case, well spotted sir, it took me a few years longer to realise that something was seriously amiss.

Bayard said...

Oddly enough, in the old days, (up until the 70's?) if you were talking about agricultural land, that is what you paid for when buying a farm, the house was effectively thrown in free.

However, I can't see how your statement about the land being thrown in free stacks up. The developer wasn't given the land for free and the builder isn't going to sell for below cost, so the eventual homeowner must have paid something for his bit of land. Also if the average homebuyer got the land for free, then homebuyers in more expensive parts of the country must have paid something for it, which implies that buyers in the cheaper parts of the country were paid to take it off the builder's hands, in which case, how did he make any money?

Lola said...

MW - not that clever really. I just sort of felt it, and there were a few straws in the wind of absolutely daft mortgages - for people on benefits say. I just decided that it felt all wrong. Also we were exposed to the appalling FSA and it was clear that they had no clue, but were engaged in a sort of nationalisation by stealth.

Mark Wadsworth said...

B, I was not talking about farm land.

Of course land wasn't "free" in the sense of zero pence. But if the whingeing Baby Boomers paid £50 or something for the land, given inflation and everything that is still f- all in the grander scheme.

As to "expensive areas" I was talking about averages of averages. But in the 1990s, you could buy new build three-bed homes in Leytonstone, east London for the same price as old ones, about £75,000.

I bought one and the insurance quote said "Rebuild cost £110,000" which was clearly wildly overstated, but broadly speaking, the land was "free".

Ten or fifteen years earlier, 3-bed houses in that area were going for £40,000 or so, whether new or old, so that must have been the construction costs.

As a self-confessed Baby Boomer, let me reiterate: we got our land for free, full stop and no back chat.

Bayard said...

"3-bed houses in that area were going for £40,000 or so, whether new or old, so that must have been the construction costs."

Howso "must"? It matters not whether you buy new or old, you are still paying partly for the land and partly for the house sitting on it. All you have demonstrated is that second hand houses were no cheaper than new ones.

Back then, if the house was in poor condition, then the value of the bricks and mortar element was almost zero. However that didn't mean you could buy it for nothing, or even £50. ISTR prices of about £10,000 being asked for derelict cottages. I paid £15,000 for a condemned 2-bed cottage in the early '90s, it cost me £30,000 to do up and was worth about £45,000 when I'd finished, which was about what it had been worth a few years before I bought it, all of which suggests that almost the entire purchase price was for the land.

Mark Wadsworth said...

B, we are talking averages, not special cases.

House prices in the UK have gone up on average by 250% in the last twenty years (i.e. times by 3.5).

CPI inflation was 50% (times by 1.5)

Clearly, a small part of the purchase price of new houses back then was greater than the bricks and mortar cost - that was the builder's profit margin, the land itself was pennies i.e. as good as free.

So a £100 house then is a £350 house nowadays, £150 is of that is cost + builder's profit. £200 is land.

If you want to argue that £10 of the original £100 was land, fair enough.

So that £10 then becomes £15 now.
The £90 bricks becomes £135 now.
So land is £265 now.

So the whining Boomers paid £15 for something now worth £265. They only paid for 6% of what the land is now worth.

So fair enough, to recognise this, we could have LVT of only 94% of site premiums/land value.

Bayard said...

"If you want to argue that £10 of the original £100 was land, fair enough."

Well, no I don't. I want to argue that about £30 of the original £100 was land. You haven't produced any figures either to say that it is £10 or it isn't £30.

Reworking your example with 30% land gives a building cost of £105 and a land cost of £245, meaning that the purchasers paid 12% of what the land is now worth.

This is a bit of a pointless argument, because whichever way you look at it, anyone who bought before 2001 has done pretty well out of land price inflation. However, saying that the land was free originally is unnecessarily over-egging the cake.

Mark Wadsworth said...

B, it was no more than £10.

If selling price was about 3 x average earnings and (re)build costs plus developer's reasonable profit margin/cost of risk and cost of capital was also about 3 x earnings, there's not much left over for the land element, is there?

Bayard said...

What are you basing your rebuild costs on? A more failsafe approach would be to use the cost of building plots. From here ( https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&ved=0CC0QFjAC&url=https%3A%2F%2Fwww.gov.uk%2Fgovernment%2Fuploads%2Fsystem%2Fuploads%2Fattachment_data%2Ffile%2F305680%2FTable_563_-_Discontinued.xls&ei=vUjZU-LKBNCa1AW_r4GYCg&usg=AFQjCNEcA469i1hbYeqC6gMpCJ2OwrKofQ&sig2=XBU5h8emngxtgTj9dFc9-w&bvm=bv.71778758,d.d2k&cad=rja ) we get that the average price for a hectare of building land in 2010 was £2,371,549. From here ( http://www.voa.gov.uk/dvs/_downloads/pmr_2011.pdf ) we get that building land in '93 is 20% of what it was in 2010, i.e. £474,310. From here ( http://www.china-up.com:8080/international/case/case/599.pdf ) we get the average build density of 25 houses/hectare, so an individual building plot in 1993 was worth about £19,000. That's hardly "free".

Mark Wadsworth said...

B, I'd argue that 3 - 3 = 0.

But if you're right and it was £19,000 in 1993, then construction costs were only £40,000 per home, index that up by 1.5 and that means that an even greater proportion of current values relate to pure land.

The all the figures must be in line - what the pre-2001 purchasers paid for land and what the current land value is; what construction costs were then and what they are now.

If you argue up pre-2001 land values, then you argue down construction costs and hence argue up current land values.

It's basic maths. So either way, the pre-2001 purchasers only paid for a tiny fraction of their current land value "out of taxed income" (glossing over MIRAS).

Bayard said...

"an even greater proportion of current values relate to pure land"

Using your example above, approx £33 of the £100 purchase cost is land, therefore the indexed up building costs are £100 and the indexed up land value is £250, which is 7.5 times. Using the figures above, we find that land has gone up 5 times, which gives a land value element of £165 out of our £350, or 47%, so depending on which figures you use, the percentage of the current land values that the purchasers paid if they bought in the bottom of the post crash trough, is either 13% or 20%. Neither prices are, by any stretch of the imagination, zero.