Rising oil prices; falling property values?
by Gavin R. Putland
(Address to the Melbourne Unitarian Peace Memorial Church, August 17, 2008.)
Thank you, Peter. And thanks to all of you for welcoming this former Trinitarian Methodist, and current Trinitarian Orthodox Christian. I’m here as Research Officer for Prosper Australia, which is Australia’s leading Georgist organization. A “Georgist” is one who believes that government should be financed out of the rental values of land and natural resources and other monopolies, rather than from taxes on productive activities. I’ve been able to establish that at least one prominent Georgist, namely Dr. H. William Batt of Albany, NY, is a Unitarian. But we’re a broad church.
Henry George, the recognized founder of our global movement, was Episcopalian. Max Hirsch, the early leading light of our movement in Australia, was Jewish. William Vickrey, the most Georgist economist to win the Nobel Prize, was a Quaker. And the people I presently work with include two Roman Catholics and a Buddhist. For good measure, our present executive committee includes at least one member from each of the Liberal Party, the Labor Party, the Greens, and (if they still exist) the Australian Democrats. So collectively we’re a broad-minded group, able to accommodate a broad range of individual rigidities.
When I was asked to speak on rising oil prices, I saw that there were many aspects of the problem that one could talk about. Some people say the recent spike in oil prices was a one-off, caused by speculators taking refuge in commodity markets. In the longer term, one can argue about whether the price rise is mainly driven by depletion of supply, or rising demand, or the intention of governments to put a price on carbon emissions, either by taxation or by a cap-and-trade system. And even if we admit that the basic problem is the finite amount of oil in the ground — which seems pretty obvious to me — we still don’t know how uncomfortable things are going to get, for at least three reasons.
* First, the effective supply of oil increases as the price rises. We’re used to the price being so low that it’s profitable to extract liquid oil, but not oil in any other form, such as tar-sand or oil-shale. But as the price rises, it becomes profitable to exploit new types of deposits and to extract a greater percentage of each deposit. Indeed, oil is already being extracted from tar-sands in Canada, and exported to the USA.
* Second, when you start exploiting a new source of oil, the technology gets cheaper over time, and reduces the minimum oil price at which the technology remains profitable.
* Third, as oil gets more expensive, people are driven out of their cars and into public transport, some of which is connected to the electricity grid and hence ultimately powered by some other natural resource, which may be renewable or non-renewable. The speed of that process depends not only on economics but also on politics. And as that process happens, it exerts negative feedback on the price of oil, by reducing the demand.
Considering all that complexity, I’d have to be very brave to offer any sort of timetable on the price of oil. But, given that the general trend is upward, I am in a position to comment on the economic consequences and, in particular, the consequences for property values.
So, what determines the value of real estate? Some people say “Location, location!” I’ll come back to that. Some people say “supply and demand”. But that’s not very helpful because supply and demand don’t just fall out of the sky; they have causes. And “demand” doesn’t simply mean what people want. It means what they’re willing and able to pay for, and how much they’re willing and able to pay. And those things also have causes. Furthermore, the value of a typical property — your home, for instance — consists of two components, and each component is subject to separate laws of so-called supply and demand.
The first component is the value of the building, or of your slice of the building, or, if you want to be technical, the value of the “improvements”. “Improvements” include everything artificial. The second component is the value of the land, or the space that the property occupies, or, if you want to be technical, the “site”. A “site” is a piece of ground or airspace, and its value includes the value of any attached rights to build on that ground or into that airspace, or to use that ground or airspace for particular purposes — in other words, the zoning. So we have the value of the improvements plus the value of the site.
Now an improvement, such as a building, is a product of competitive effort. So its value is limited by the replacement cost, less an allowance for depreciation. Its value can fall below that limit, for example if the building is no longer appropriate for the location; but it can’t rise above that limit for any considerable time, because if it did, you’d rather build a new building [than buy the existing one].
Now I can explain the expression “Location, location!” The locational value is included in the site value, not the value of any buildings, because the values of buildings are constrained by construction costs, whereas a site has a location, and therefore a locational value, even if nothing has been built on it yet — indeed, especially if nothing’s been built on it yet, because that means we’re still free to build whatever makes the most profitable use of the location.
At this point I can also draw the first conclusion about the effect of rising oil prices on property values. And it’s not very surprising. Oil prices feed into travel costs, and travel costs are buried in the prices of both building materials and building services. Furthermore, some materials used in connection with buildings are petroleum products, making their prices doubly sensitive to oil prices. That’s unsurprising because we’ve grown used to the idea that oil prices feed into other prices; so we expect them to feed into construction costs, hence the values of buildings.
But that argument doesn’t work for sites, because sites don’t have a replacement cost, because they don’t have a production cost. Sites are land and space, which come to the community as free gifts of nature — or, as we monotheists say, gifts of God. That raises the question: How did private individuals and corporations get the right to charge people for the use of sites? To which Henry George answered: They didn’t! Nevertheless, people are willing to pay for sites because sites are limited in supply, and because some of them are more desirable than others. For an economically rational user of land, the rental value of a piece of land is as follows (and I quote):
… the difference between the advantage of having the use of the land in question and the advantage of having the use of the nearest other land the use of which can be obtained by mere occupation without making payment to any person for such use.
(Unquote.) The quote comes from the “Elementary Property Law Bill” of 1890, which was unsuccessfully sponsored in the Queensland Parliament by Samuel Griffith. Yes, him. Yes, he was a Georgist. Which, by the way, is another illustration of our religious diversity, because he was also a Freemason. Yes, the first draft of the Constitution of this country was written by a Georgist. And he couldn’t have been alone in his views, because that Constitution still requires that all the land in Canberra is owned by the Commonwealth and available on leasehold only. (And if only the ACT government charged anything like market rents, there’d be no need for stamp duties or payroll tax in the ACT. But perhaps I digress.)
To paraphrase Griffith: The rental value of a site is the advantage of using that site, rather than using the best alternative site that can be had for nothing. That law was first stated in about 1809 by David Ricardo; so it’s called “Ricardo’s law of rent”. If all the sites are taken, so that none can be had for nothing, then we have to generalize the law by saying that the value of a site is the value of the cheapest site plus the advantage of using the site in question rather than the cheapest site. Ricardo’s law is the special case in which the cheapest site happens to be free.
One thing which Ricardo didn’t notice, and which was eventually explained in 1879 by Henry George, is that Ricardo’s law and its generalization both follow from the mobility of labour and capital. Because workers can move in pursuit of the best deal, and because investors in buildings and plant and equipment can invest their money wherever it yields the best return, the net returns to labour and capital (for given levels of skill and risk and inconvenience) tend to equalize across the country. But those returns are “net” of the costs of locations — in other words, site values.
If the optimal application of labour and capital to a particular site yields a return in excess of the going returns to labour and capital, the owner of the site can demand the excess in the rent or price of the site. So site values are the great equalizer between users of sites — but obviously the great discriminator between owners of sites, and between owners and non-owners.
When Ricardo considered differences between sites, he was mainly interested in the fertility of the soil. But we know that location also matters. And so did Adam Smith. But the first attempt to quantify the influence of location was made in 1826 by the German economist Johann Heinrich von Thünen. He said: Let’s ignore variations in soil fertility (because we’re studying locations, not soils.) And let’s assume that we have a state with a single dominant city. (Sound familiar?) And let’s assume that for whatever industry we’re in, this city is our sole market, and our sole source of supplies of any raw materials or capital equipment that we can’t produce ourselves. (Actually I don’t think von Thünen considered supplies; but I threw that in because it adds a bit of realism without changing the nature of the result.) And let’s assume that our transport cost for each dollar’s worth of sales is proportional to our distance from the city, as the crow flies. (In other words, let’s ignore roads, railways, rivers, and topographic features that might affect transport.) Because transport costs obviously increase with weight and bulk, that proportionality factor depends on how heavy and bulky our product is, relative to its value. Hence it depends on what industry we’re in.
Now, the rent per unit area that we can afford to pay for land right next to the city — in other words, the return on using that land, over and above the return that we need on labour and capital — also depends on what industry we’re in. For land further from the city, the rent we can afford to pay is reduced by a margin proportional to our transport cost, which in turn is proportional to distance from the city.
So, if we draw a graph showing the rent per unit area that we can afford to pay, vs. distance from the city, we get a straight line, which starts at a certain height on the vertical axis (indicating what we can pay right next to the city), and has a negative slope (indicating our sensitivity to transport costs). Different industries have different starting heights and different slopes. But for any particular distance from the city, a certain industry offers the most rent and therefore gets the land; and that industry prevails over a certain range of distances. So, when we draw a map showing which industries exist in which locations, we see a series of rings centred on the city, each ring corresponding to one industry. Those are called “von Thünen rings”.
Now, in the light of that long exposition, we can start explaining how oil prices affect site values. The slopes of all those straight lines on the graph are proportional to transport costs, most of which are directly related to oil prices: if oil prices rise, the lines get steeper. And because they might get steeper in different proportions, because different industries have different transport modes and some are constrained by weight and others by bulk, the industries that win the competition for sites might not stay the same. So the von Thünen rings shrink, and maybe some rings drop out, and maybe new rings appear. But most obviously, because those downward-sloping lines get steeper, land values fall; and the further you go from the city, the more they fall.
That may come as a surprise, because we’re used to the idea that oil prices feed into production costs and hence into other prices. But the crucial point is that the value of a site isn’t a production cost. It’s a “surplus”. It’s whatever’s left over after production costs have been met. So the correlation between production costs and site values is negative, not positive.
Now let me mention just a few of the factors that complicate the simple von Thünen pattern.
* First, there’s a transport system which makes it cheaper to transport goods along certain lines. Those lines form radial patterns and tree patterns, which distort the rings into corresponding patterns. And if some of those transport routes don’t use oil, then a rise in oil prices won’t directly increase the slope of the rent graph along those routes, although it will increase the slope as you move away from those routes.
* Second, a city is not only a market and supply centre in its own right, but also a gateway to other markets and supply centres. But any coastal port is also a gateway, which means that site values tend to fall not only as you move away from cities, but also as you move away from the coast. So, as well as rings centred on a city you might get bands or belts running more or less parallel to the coast.
* Third, there are multiple cities, each of which can support its own pattern of rings.
* Fourth, the von Thünen theory doesn’t explicitly allow for limited capacities of local markets and supply centres. Allowing for these things presumably reduces the sizes of the ring patterns around small cities, especially small isolated cities.
* Fifth, a city isn’t concentrated at a single point. It spreads over a geographic area, like any industry within it or outside it. But, as you move further from the city centre, you tend to be moving further from markets and supplies. For example, if you’re a wage or salary earner, moving further from shopping centres means moving further from supplies, and moving further from your place of employment means moving further from the market. So the von Thünen theory predicts, quite correctly, that site values fall as you move further from the city centre. And it further predicts that if transport costs rise, the “site-value gradient” will increase; that is, site values will become more sensitive to distance from the city centre.
* Sixth, a large city tends to have multiple suburban hubs — including former satellite towns that were gobbled up as the main city expanded. And as transport costs increase, the importance of a nearby suburban hub will increase relative to the importance of a more distant city centre. This will tend to break up a large city into smaller functional units. On a map, or on a satellite photo, it will still be one continuous built-up area; but in terms of transport and commerce, it will be less like one big city and more like multiple smaller ones.
Now, the increase in the site-value gradient doesn’t quite mean that site values are going to fall further in outer suburbs than inner suburbs. It only means that the negative influence of high oil prices is going to be greater in the outer suburbs. But that influence is superimposed on many others, including population growth, general economic growth, technological progress, and investment in infrastructure, all of which tend to increase site values.
Also remember that while higher oil prices have a negative effect on site values, they have a positive effect on the other component of property values, namely building values. But here I think the negative effect is dominant. If you build a house in a certain location, all the building materials have to be transported to that location only once, and the building workers have to commute to that location only long enough to get the job done. But if you then live in the house and commute somewhere else to work, you’re going to make that trip, and pay for that trip, 5 days a week, 48 weeks a year; and that’s going to have a profound effect on how much you or anyone else will be willing to pay for the privilege of living there.
So a complicated argument leads to a fairly clear and simple conclusion. While property values depend on many things besides oil prices, the influence of higher oil prices on property values is going to be negative, and that negative influence will tend to be greater as you move further from city centres, or further from major suburban hubs, or further from public transport routes that don’t depend entirely on oil, or further from the coast.
I thank you for listening…
Appendix (not included in address)
Von Thünen’s equation was
R = Y(p-c) – Yfx
R = rent [e.g. $/year/hectare]
Y = yield [e.g. tonne/year/hectare]
p = price [$/tonne]
c = local cost of production, including normal returns to labour & capital [$/tonne]
f = freight rate [e.g. $/tonne/km]
x = distance from city [km].