Now in the unlikely future event of a real schism with two persistent forks, it doesn't affect current investors. They can sit tight and whatever happens their money is safe. For new investors of course they have to make a choice (or hedge between them), but that is natural given there was a real schism in the economic majority, which suggests there were good reasons for it. I don't think hardly anyone is going to invest in a smallblock fork versus a reasonably larger block fork, for example. If we assume a major schism where investors support both, which seems very unlikely, then there was probably a good reason for it.
This is my argument about say increasing the block size. There seems to be deep division about whether this is a good idea or at least whether some of the specific proposals of how to do it are a good idea.
Maybe you are right and trading would resolve the issue quickly if people had to put their money where their mouth is rather than just make a lot of noise over it. But if not, then exchanges, wallets, payment processors, etc. could be set up to deal in hedged baskets which consist of one of each coin traded together. Those who did not want to bet on one or the other could just use the basket until the market sorted out the question one way or another.
I no way is this a suggestion to simply let the chain fork as much and as often as anyone wants without the necessary infrastructure in place at both the exchange and wallet level to deal with it in a non-disruptive manner.
This suggests a kind of graduated strategy for forking. As opposed to the old strategy of...
1) Musing on fork possibilities
2) Debate of concrete proposals
3) Consensus
...there is a new more gradual strategy:
1) Musing on fork possibilities
2) Debate of concrete proposals
3)
Rough consensus
4) Market arbitrage
The second strategy seems especially useful when getting full consensus is hard or is taking too long and time is of the essence. If blocks are filling up and transactions are getting delayed, some larger-capacity altcoin is leeching market share away by the day, and the media is having a field day, I imagine getting consensus to raise max_blocksize gets much easier. But in those cases where a straw poll isn't enough, putting it to a "market vote" is the final test.
It's only appropriate when every other means of assessing consensus leaves it indeterminate which proposal will win.
