I have been convinced of the benefits of LVT (which has been LibDem policy before, when it was known as Site Value Rating, or SVR) for some time. But I don't think it is suited as a replacement for general taxation.

What it is ideally-suited for is the funding of capital-intensive infrastructure projects, let us say a new Tube line.

At the moment the capital for infrastructure investments has to be raised on the capital markets and is added to the PSBR. Most of these investments do not show a conventional positive rate of return, but they are beneficial to society, and the usual result is an increase in the value of properties which benefit from the investment. This increase in value is due to the investment, but the problem is that capital-intensive infrastructure projects effectively result in a transfer of value from the general public to the particular property owners affected, which is inequitable. LVT addresses this inequity.

Now, suppose that instead of raising the capital for these projects by a fixed-coupon bond, it was raised by a bond the coupon of which was calculated in terms of the LVT yield on properties in the affected area. Lenders would know that this would increase as a result of the project, so they would be willing to lend at a lower initial rate. Furthermore, these bonds would provide a convenient way for property-owners to hedge the increase in LVT that the infrastructure would entail.

The mechanism could be taken further and used as the metric with which to pay private-sector operation and construction partners: the private sector operator of an infrastructure concession would be paid the proceeds of a site-value based "penny rate" (or a fraction thereof) on the land in a defined area, for the duration of the concession. The partners would then raise the capital themselves in whatever way they saw fit, but they would have the security of a future income stream which could be mortgaged to do so.

Deciding what "the affected area" is gives problems, also deciding how much of any changes in property values is attributable to a particular piece of infrastructure. It works if (like California) you are irrigating a piece of virgin desert. It doesn't work nearly as well in somewhere like London.

That's why a general tax on land values, paying for all publicly funded capital investment in a whole jurisdiction, is simpler and fairer.

You can still use the bond markets however (not that I'm a banker) but in a more general way. In the UK, an Olympic Fund levied on the whole of GLA could fill the gap that Gordon has in paying for transport investment.

First, the bonds I'm suggesting would be limited by local-government area or similar, so a Tube bond might be levied on property on the whole of London. Your idea of an Olympic fund for London makes sense too. Those who benefit most from the Tube investment will pay more, because their properties will increase in value more as a result of the investment (as they would with a national scheme).

The bonds, however, should be closely linked to the LVT. They aren't "x% per annum" bonds, but (say) "Greater London 1p LVT bonds"; what the bondholders would get each year would depend on the value of the land in the area concerned. The bigger the impact of the project on land values, the more they would get.
Authorities could also try issuing more highly-geared bonds - differential LVT bonds - where the coupon, rather than being a flat proportion of the LVT yield is a proportion of the increase in LVT yield over a baseline.


The point is that this puts more of both the upside and the downside risk in the capital markets not in the public purse. If the project is a failure (and some can be) future taxpayers don't get landed with paying for it. It provides a way for the public sector to raise something approximating to equity finance. This is also an argument for paying some or all of the contract in bonds rather than cash.

The yield of a proportion of the LVT raised by any area is hypothecated to debt service, but in an transparent and trustable way. Large amounts of long-term capital could be raised cheaply to finance major infrastructure projects without running into the difficulties presented by the public-sector's predominantly cash-based accounting system.