I've done a time-series regression of amount of track within a 40km radius of Holkham Hall against land rent, controlling for the price of wheat and cattle. I deflated the rent and the commodities to adjust for changes in the cost of living. The result is statistically significant, and I'd love to show you the output but can't work out how to paste it into the blog. Track and cattle have positive signs, but wheat is negative. Why should a drop in the wheat price result in increased rent..I don't know yet! I'll get there. The positive sign for Track is what we had expected to see: more track means increased rent. The savings are being extracted by the landowner. Recall that the equation for locational rent is

where m is the market price of the commodity, c the cost of production, E the yield, f the cost of transportation per unit distance and d the distance. As the distance increases, the right hand side gets smaller, so the rent gets smaller. Eventually the rent would be zero right on the edge of the cultivatable land. By increasing the amount of track, in effect the distance is getting smaller...so the rent goes up.

Encouraged by this, we're going to build a larger dataset. Malcolm is calculating the track for three more estates: Petworth, Thorndon and Dalemain. A graph of their rents is here:

You can see that there is a bump in the rents in the period around 1850----what a coincidence! Once we have the track data in, I'll do the same type of regression, but this time it will be a panel-data longitudinal regression. This is a very powerful technique, which I'd urge you to learn if you see the chance.

Subscribe to:
Post Comments (Atom)

## No comments:

Post a Comment