Tuesday, July 10, 2007

Based on GDP per Kilometer, Property May Not Be as Overvalued as You Think

Property, it is assumed, always goes up in the long run. This is based on the assumption that there is a limited amount of land, yet the population will continue to grow, therefore housing prices will forever trend upward.

However, if that was the case then countries like Ethiopia, Somalia, Sudan, etc., that have significant populations would have high property values. Alas they don't since they are constantly plagued by war, famine, plague, etc. However, they are also plagued by another disease; the lack of economic growth.

While I'm sure at some level population does correlate with higher property prices, a better determinant of property prices is the amount of wealth created from that land. The more wealth that can be produced per square mile of a particular piece of land the higher its underlying rents would be and therefore its value. Thus I crafted a new ratio;

GDP per Square Kilometer (I had to use kilometer because the CIA World Fact Book has the countries' area in kilometers)

Real GDP per sq. km. in the US has gone from just under $200,000 almost $1.3 million today. As technologies have advanced, managerial efficiencies invented and employed, we here in the US are able to squeeze out almost 7 times the amount of wealth from our land per square kilometer than we were just 50 years ago (quite identical to farming yields on a per acre basis).

It is this increase in wealth that we can extract from each square mile of our land that has truly increased our property values. However, combine the two, high levels of the production of wealth with high populations, and you get property that is most highly valued; cities.

It is no coincidence that New York, London, Hong Kong, Singapore, Tokyo etc. etc. have the highest property values in the world because not only do they have some of the largest populations, but they are also centers of commerce where disproportionate amounts of wealth are created.

However, GDP per square kilometer also provides us with a tool by which to gauge property values. In theory as more and more wealth is squeezed from a particular piece of land, the value of that land should increase proportionately. Therefore the ratio between GDP per square kilometer and the value of that land should remain constant. Introduce a new measure;

Median Housing Prices divided by GDP per Sq. Kilometer.

In theory the average price of a home in the US should increase in sync with GDP per Sq. KM resulting in a constant ratio. However, that has not been the case;

Average (mean) housing prices were originally 28% of the wealth produced on a per square kilometer basis in the US. This has trended downward to 19% in 2001, only to recover to 21% in 2006.

In other words, the value of our land is not keeping up with the wealth that is produced by it. And although property prices did seem to recover in the latest housing bubble, they look set to trend back down towards the 19% mark (coincidentally implying a 10% over-valuation in the housing market).

Therefore there must be a macro-economic variable driving this ratio down. Without additional research and left to guess, I surmise it is the transition from manufacturing to services that is disconnecting the market value of the land from the wealth that is produced from it. Services require less land than manufacturing. Goldman Sachs has produced ever increasing wealth from it's meager .9 square kilometers of land it owns in New York by offering ever more profitable services, while GM has barely produced squat with its arguably scores of square kilometers of land that its factories and facilities sit on.

Alas I'd be curious to see if my theory would hold for China which is experiencing a manufacturing boom.


Bill Gilles said...

I just I have to say that this is really impressive and unique analysis. I appreciate you going out of your way to challenge your own assumption of a housing bubble and why it may or may not exist. Keep it coming!

Anonymous said...

Thank you.

Anonymous said...

Hey, fascinating analysis. I'd like to see you stick the lefties by staying away from the metrique system, though. Google square kms per square mile and convert your data. Quick, easy, and nonmetric!

Anonymous said...

I agree with Bill, this is very good and crisp thinking. I would agree with your hypothesis, at a glance, that a macro-level phenomenon pushing prices down is the transition from a manufacturing to service economy. In concert with this, I surmise that the spread of technology, making it possible to create wealth (i.e. communicate from offshore, limiting the necessity of central offices) is having some effect on pushing real estate prices down. Perhaps the Internet is "new real estate"?


Captain Capitalism said...

Yes, it is no doubt the transition from manufacturing to sales, but there are also other factors involved, technology no doubt being one of them. The internet or a web site, takes up no space whatsoever bar a server, yet can yield an amazing amount of GDP (look at Google's servers) all lessening the connection between wealth production and the land required to produce it. It's why I would be very curious to look at China, Vietnam, and to a lesser extend India to see if they would be experiencing a different median property value to GDP/Sq.KM value.

Anonymous said...

So, I was going to do this graph, and as an amateur economist I wanted to see if it did have any relevance.

I came across a chart on some leftist peak oil site that had they realized the chart was from the Economist, they might have thought twice about posting it (http://tinyurl.com/2tauzy).

The leftist site droned on about how American public transport is the worst, except in highly blue cities like San Fran and NYC. I responded with the like that the NYC subway system has a budget larger than the GDP of the Bahamas, and that it runs a deficit such that over 50% of its operating costs must be covered by taxes instead of user fees, but that's beside the point.

Now, of course, it's a blame America first crowd there, and oh we suck gas and all, and thought to put together a chart that dealt with GDP Produced per Gallon of Gas consumed. This shows the US to be not too great with ~$100 of GDP per Gallon compared to Japan or the UK, and then I saw the chart on GDP per Kilometer and thought, what I'm missing here is the population density.

Now, what I'm asking is, is GDP per Gallon per Population Density in People per KM^2 a bit over esoteric? Or is there a better line here? It definitely adds more to the cost of goods sold if you have to transport your goods thousands of miles to reach your markets, and with the US with a 31 people per km^2 versus Japan with 337 and UK with 383, the US would require more petrol to get those goods to the same amount of people compared to more dense countries.

Now, when doing the numbers, it definitely benefits sparse countries (i.e. Australia, Canada, US and Brazil) but really punishes Japan and UK, which is why I don't know if people per square km is that valid of a number.

Proud Owner of an Autographed Captain Captialism photograph.

feenpup48 said...

very nice theory & facts to back it up! The only problem is that there is a significant decline in home values in the state I live in, Massachusetts. This is occuring while gdp is increasing. What you fail to discuss is that if wealth is concentrated at the top, the broad based real estate market can't sustain high prices. To sustain a high-priced market, AVERAGE income MUST increase across all or most sectors. Thus gdp per km is irrelevant...sorry boss!

Joao Coelho said...

I would say that property prices in big cities will tend to decrease, as technology allows many jobs to be equally done whether you are in downtown NY, in a Newark suburb or in a a small town near Bangalore.
This means many companies and population will move/are moving to cheaper areas. Nevertheless, the increase in property value in a suburb will never compensate the loss of property value in NY.
Therefore, in the long term average prices will go down.
As the Internet becomes ubiquitous, the logic of needing to be physically/geographically near your work colleagues, suppliers, partners or clients to be able to create wealth will tend to disappear. At the same time, as energy prices go up, people will try to waste less time and money commuting to work.