I happened to be in Chester yesterday when they were holding the memorial service for the Duke of Westminster. The security was pretty stringent and there was clearly a major logistical operation taking place. There were definitely strong echoes of British feudal history. Probably not coincidentally, there was a piece in the FT describing how many of the traditional estates, with Grosvenor at the vanguard, have transformed themselves from feudal miscellanies into modern institutional real estate investment management organisations.
The traditional estates seem to be trying to reconcile the perennial real estate investment dilemma problem of specialisation v diversification. Whilst they may proclaim the largely un-evidenced synergies of geographical clustering, some have been implementing a strategy of geographical, sometimes global, diversification. I suppose that you can do both. If you have enough money, you can own a number of clusters spread across many markets and try to improve risk-adjusted returns through long term relationships and investment to ensure significant benefits of pro-active management are not captured by free-riding neighbouring owners.
This is basically the shopping centre argument whereby a single owner-manager is able to co-ordinate and optimise the retail environment and maximise (minimise) positive (negative) externalities. When a single owner controls a city quarter, similar incentives are created and similar strategies can be introduced. I know that Patricia Canelas at UCL is working on this topic. There’s a plausible case here and many investors prefer specialisation since they can diversify themselves. But I haven’t seen any convincing published empirical evidence of major performance effects yet.