Strategies for not losing money? Well, generally, there are three things that can cause someone's bank account to hemorrhage.
First is unit support costs. Press F3 or open up the Military Advisor screen, and he'll tell you how much GPT you're spending on your troops, if any. This is especially important in a Republic - each unit over the small amount of freely supported units costs 2GPT as opposed to 1GPT. Despotism, Monarchy, Fascism and Communism all have large amounts of free unit support, so for the most part you should be safe, although it's still possible to go over. Republic has a small amount of unit supports, and going over costs alot extra. Democracy has no unit support, but each unit only costs 1GPT as opposed to the Republic's 2GPT. Feudalism is a unique government, though, where you get lots of free units from having alot of small towns as opposed to the traditional setting where you get lots of free units for having alot of big towns.
Second is improvements. One thing that greatly increased my skill, especially when playing as a Monarchy/Fascism/Communism, is to not build every improvement in every town. This is especially important with Factories and Coal Plants. Unless a town has an already good shield production (Say, 25-30) AND it can handle losing 6GPT, it's generally a bad idea to put a Factory and Coal Plant there. Just remember: cities that can't grow above 6 don't need aqueducts or hospitals, cities with low shield production don't need Factories/Coal Plants, cities with small populations don't need Cathedrals or Colosseums, cities with low gold income don't need Libraries/Universities/Research labs, cities lacking any significant shield production don't need barracks because they won't be building troops, and cities over population of five don't need granaries. It is a good idea, however, to have a Temple and a Marketplace/Bank/Stock Market in all cities, as well as building Harbors and Commercial Docks on all coastal cities.
Thirdly and probably most importantly, you need to be able to use the science slider better. Even if you're taking a small loss in GPT when keeping it on, say, 60%, that doesn't mean you can't walk away with more money. The thing to do is to check whether or not you can get the tech in the same amount of time with less money put into it. Usually, around when you have four turns to go, you can afford to lower the slider by 10 or 20% and still get that tech in four turns. Just keep checking once the tech approaches completion - almost always can you squeeze out some extra gold without slowing down the research time. This will allow you to run at, say, a -6GPT for the first seven turns of research, and then turn it up to a +12GPT for the last three turns and still get the tech in the predicted ten turns. It's a very good way to squeeze a little extra money into your treasury.
Hope this longwinded post helped you some.