Killer arguments against LVT, not (42)

Whereby I finally get round to dealing with IanB's claim that a tax on land values is a tax on imaginary values:

... you're taxing something imaginary; that is the hypothetical rental value of land which is not rented, or the hypothetical sale value of land which is not for sale. If you object so strongly to profit from land, why don't you tax rents or sales of land? Then you can still capture all that nasty unjustified rent profit you hate so much. Why tax something hypothetical instead of a real transaction? Why tax estimated "value" instead of actual values measured by the economy during rental or sales transactions? Answer me that question Mark, please.

My initial response was thus:

I believe in substance-over-form. Let's imagine a tenant agrees to buy the house he is living in from his landlord. It is the same person living in the same house, and probably paying a mortgage to the same bank. Can you please explain why you think it desirable that, from that transaction onwards, the owner of that land no longer has to pay any tax? Do you not think that at the very least this is a massive distortion, and indeed tilts the playing field away from the much maligned landlord?

To which he replied:

Value only exists during transactions, Mark. It doesn't exist at other times. You do realise that?

Well, no, I don't accept that, as I hope my simple example showed.

1. Never mind, if you can be bothered to read this thread, it is pretty clear that even between ourselves, Land Value Taxers have never really agreed how the tax would be calculated (even though I have strong views on how to get the ball rolling).

2. The answer is in fact quite simple provided you remember The Golden Rule: The tax should only be on the total rental value above beyond the rate of return on the bricks and mortar. There is no magical formula; it requires very little by way of maths or even expertise, and the tax can be set in the same way as any monopolist sets his price, i.e. by trial and error.

3. By definition;
i) the only reason why the selling prices (or rents) for physically similar houses in different parts of the UK differ so wildly is because some locations are more desirable and others less so.
ii) residential property is very lightly taxed in the UK (compared to incomes and production, which are far too heavily taxed),
iii) a large part of the location rent is capitalised into selling prices.

4. Therefore, if the tax rate were set correctly in each postcode sector, we would observe that capital selling prices of land and buildings are equivalent to the replacement cost or value of the bricks and mortar; and that the capital selling prices of physically similar properties would be very similar across the UK (with the exception of particularly run down areas, where the selling price may be less than the rebuild cost).

5. Seeing as the bulk of UK properties fall into a dozen or so broad categories (three bed semis built between 1920 and 1960; Victorian or Edwardian terraced houses; office blocks; industrial units etc) it would be quite easy to work out what the rebuild cost/value of each of these is (e.g. by using the ABI rebuild cost calculator and knocking off a third because they exaggerate), it can't be rocket science to continually monitor selling prices (which is what HM Land Registry and the DCLG already do, of course) and adjust the tax rate in each area up or down annually.

6. We can simplify all this even further by just taking the average build density in each area, e.g. 0.3 for most residential areas, which means you can build a 120 sq yd house on a 400 sq yd plot, multiplying that by an average bricks and mortar replacement value of £720 per sq yd, so the selling price of a house on a 400 sq yd plot 'should be' 400 x 0.3 x £720 = £86,400.

7. So if older three-bed semi-detached houses in an area sell, on average, for more than £86,400, then the tax per square yard must be 'too low' and gets hiked a bit in that area; if they sell for less than that in another area, then the tax is 'too high' and the tax is reduced. Thus by pure trial and error, the tax would pretty much = the rental value of the location. I happen to know that there is a colossal disparity between different parts of the UK, in some areas the rental value may be £1 or less per sq yd per year (for agricultural land it is more like 10 pence, which is part of the reason why it is not worth taxing) and in prime central London it is over £1,000 per sq yd per year.

8. The willingness (or otherwise) of people to pay the tax gives us a continuous stream of transactions (dealing with the second limb of IanB's objection), which indicate the rental value of the plots of land themselves; if you decide to stay living somewhere, you are tacitly accepting that the tax is not 'too high'. If the tax were 'too high' in an area, then selling prices would fall below rebuild cost/value and the tax drops again. It's like a glorified part-rent-part-buy scheme, is all.

9. And of course, the aim of all this is to replace taxes on incomes and production; if and when those have all been eliminated, any surplus the government makes is dished out as a Citizen's Dividend (or you can be a bit more paternalist about this and dish it out as earmarked payments, like old age pensions and school or health vouchers, different topic).

10. And yes, such a tax would be wildly unpopular with some people, but that is yet another advantage - 96% of respondents to this Fun Online Poll said that they preferred in-your-face taxes to stealth taxes, for the simple reason that the government finds it harder to raise money from in-your-face taxes.