Just watched an episode of "Shark Tank" in which a guy and his wife sought funding for his dog training academy. He provides a personal consultation, finds the client's needs, and hand-picks the dog for the client. He then trains the dog in a number of security techniques. The lowest price for his service: $20,000. The highest price is $40,000. Bear in mind that he only picks from four breeds, including some that are in use by the US Army and US Special Forces. So the dogs themselves aren't exactly cheap.
But for a pound puppy, I'd say that it's not unreasonable to ask about $200 to cover the pound's expenses, licensing, tracking chip (RFID tag), and some starter gear like food and training pads. Afterward, the percentage-based lifestyle increase makes a lot of sense.
Obviously there are a LOT of variables, even when talking about just dogs, and some cats. When you start to get into some of the more exotic non-paracritters the costs can be truly staggering. This is part of the reason why the books have always given the GM a LOT of leeway when making these kinds of decisions. It's a lot like the thread about buying bicycles and other items that aren't in the book. The only really good answer is this: GM call. The GM should be making at least a semi-educated decision, and the GM should be making this decision with the full understanding of the player's intentions and goals in mind.
For instance, if the player wants a dog now but plans on making that dog into a biodrone, the GM should probably explain that biodrones can't be made out of every specimen and sometimes increased costs are necessary to get just the right dog. Without that information, the GM is making a decision that will come back to bite someone in the ass - perhaps literally.
