There have been a couple of threads in Ship Model Forum about this very subject recently (this past month). Some of them quite heated. The consensus there is that while warships are undoubtably more popular, there is quite a bit of interest in non-military ships. Especially passenger liners. But since very few of these even resembled each other, much less used the same plans, it is harder for a model manufacturer to get a good return on a model. With a model of the Titanic, someone can make one of its sisters. But how many of these liners had sister ships? Most were one-of-a-kind. Some had several configurations that could be made, such as Troopship service, which would give the manufacturer a little more leeway to make multiple models from one set of basic molds. But most warships had several sister ships (some, dozens!). The main subject on SMF was the Normandie, and why no plastic company has made one in something larger than 1/1250 scale (I think it was Dragon that made that one). There was a lot of argument about the reasons why and why not. Basically, it comes down to whether the manufacturer can sell enough units to pay for doing the research, and mold-making, and designing, etc. I think someone mentioned a figure of 10,000 units as a minimum break-even point. If they only make 5000, then the price may be to high to actually sell that many. But if they price it lower and make 10,000, then they may be stuck with a couple thousand leftovers. Its a delicate balancing act.