Death and geography

I once predicted the demise of the mainframe – for a few minutes.  I’d been running a pan-European survey project for a major computer manufacturer into the market for all sizes of computers for the coming seven years.  Extrapolating from history, mainframes would be dead within seven years.  My truly-brilliant boss told me that “things don’t work that way”, and saved me from myself.  We did the survey back in 1984 (spooky!).  Mainframes are still with us.  They look a bit different today, though.

On the same basis, expectations of the demise of European exchanges are probably exaggerated.  But exchanges may look different in a few years’ time.

At the moment, dealing rooms are fighting a geography battle, and using technology to do it.  They are distant from the execution venues that they want to trade on, and they are using technology to cover the distance faster.  Latency is the keyword.

There’s that old joke – the American tourist who’s lost in the heart of Ireland, and who asks a local for the way to Dublin.  The local’s response was “If I wanted to get to Dublin, I wouldn’t start from here!”  At the moment, dealing rooms are trying to get to execution venues by starting from somewhere else – somewhere where the execution venue isn’t.

Once the different firms’ technological edge gets evened out, and everybody is using the darkest fibre and the fastest processing, maybe the next competitive step will be to change the geography – to bring the execution venues closer to the dealing rooms.  Dealing functionality and execution venues would then cluster together as physically close as they can get.  They would share the same geography.

Exchanges that are in the wrong geography would reduce in size and importance.  But they are unlikely to die.  Things don’t work that way.