On top of all the above reasons is that there is no compulsion to retire them until such time as there are live overhead conductors in the way of safe running. Leave them where they are until there is no room for them or no place to run. Don't want to rush these things as the sooner you do anything, the sooner you do something else, which in the long run means you are spending money faster.
The one problem with this logic is that keeping them going could well be more costly than "spending money faster" - the first "do anything" action is to stop throwing money old trains built during the Cold War.
Eleven EMUs have now entered service, which leaves only two more being needed to cater for the Seaford extension and fully replace the 2000 fleet with the same amount of margin in the fleet to cover for unreliability. With that in mind, it's surely now an option to stop spending money on the 2000 fleet and let attrition take care of the retirement schedule - unless they are so bad that the combination of the available inventory and part stripping from failed units won't be enough to get even just a couple through to early next year when 4013 enters service, but if that's the case they sound more like liabilities rather than assets.
If a couple of 2000s can be nursed through until 4014 and 4015 enter service, that will create slack in the DEMU fleet which will allow the DEMUs to have a more generous preventative maintenance program started. Unless that mythical acceleration in EMU production to one unit per fortnight (as was promised in January 2013, April 2013, August 2013, November 2013 and a couple of times this year) comes before then, that time should be around February-March next year.
Whether some will need to be retired before they are run into the ground will depend on whether there anyone makes a good enough bid to buy any in serviceable condition.