If miners can make a positive return by producing larger blocks ... they will.
How? By forking the blockchain?
Eventually that might happen if the limit isn't raised. But hopefully that's not going to be required.
They want money, more money is always better.
And they'll get more money by not colluding to artificially limit supply. Otherwise there will either be a fork, or there will be an altcoin, or there will be an anti-trust lawsuit, or something will give.
(That is, assuming you're correct in the first place. It's not actually true that all miners care only about money.)
However tonight you could flip the network to 1 GB blocks and ACTUAL block sizes aren't going to change much.
They'll change though. Eligius will soon start creating larger blocks.
Nobody is making 1MB blocks. Nobody. Not a single block in the history of Bitcoin is 1MB. So the "limit" isn't a limit. Miners are choosing their own parameters which result in blocks less than 1 MB.
It would be like the fastest car in the world is 180 mph and there is too much traffic so the government decides to raise the speed limit from 500 mph to 2,000 mph think that is going to have any material change. The limit will be raised eventually I have no doubt but right now the constraint on tx volume isn't the 1 MB limit. That is pretty obvious when there are no 1 MB blocks. The constraint is on the economics of mining. Eligus for example makes some of the largest blocks on the network. Routinely over 500 KB but they also include no free transactions.
Eligus couldn't make a 1MB block right this second even if they WANTED TO because there aren't 1MB worth of paying tx waiting for a block. So how would raising the limit to 2MB, 5MB, 10MB change anything to that equation?
Case in point here is a recent Eligus block.
https://blockchain.info/block-index/444092/00000000000000029fd11f8e23b450749807f78ab4aa789b764cd10ea7062e59780KB
1280 txs.
It couldn't be any larger because it includes 100% of the tx which met Eligus inclusion requirements.