Late Nite Note On Averages And Extremes

Banking analysts at Societe Generale wrote last night, my emphasis in the last sentence:

The myth of the “average” LTV. We believe the average LTV is a largely useless measure when estimating potential losses. A simple example shows that, despite a low average LTV, banks can still lose money. It is not the average that determines losses but the extremes.

Then, Kenneth Rogoff had an article at Financial Times yesterday, where he wrote:

In China today, the real problem is that no one seems to have very good data on how debt is distributed, much less an understanding of the web of implicit and explicit guarantees underlying it. But this is hardly a problem unique to China. Even as published official government debt soars, huge off-balance-sheet guarantees and borrowings remain hidden for political expedience around the world.

So, focus on extremes, and forget averages ...