Okay, I'll try to spell it out then.
Imagine that you're in a multi-player game as England, and have managed to set up a super-long trade route from East Asia all the way back to London. You have a majority of power in each node so that, as you say, you're getting more income from each merchant boost than you're losing to competition. When it gets back to London, you have more money than you know what to do with.
But your rivals in Europe have also been getting loads and loads of money just from the drips and drabs they've siphoned off from you, and they know exactly what to do with it. Your fleet is light ships scattered all over the world. Theirs are heavy ships and transports which — before you know what's happening — have blockaded and invaded you, forcing you to hand over all the outposts you relied on to maintain that world-spanning trade route you invested so much in.
If you'd just collected everything earlier in the route, where you had a total monopoly and before any other Europeans had had a chance to get to skim a penny off you, you would have made a little less, but this would never have happened.
Now, that specific scenario is unlikely, but the general point holds true: in a competitive situation, you do not want such large amounts of money leaking to your rivals. With human opponents (or just more competitive AI), such super-long routes are not going to be feasible. So they're a bad habit to get into.
Lots of posters seems to be very excited over a trivial single-player exploit of the merchant boosting effect at the moment. But if patches do not change it, then the best mods will. And it would not work in a competitive environment anyway. And you can make plenty of money without doing it, so replying to every person asking for advice with trade by saying, 'What you need is this ultra-route from California to Antwerp' is not very helpful.