There exists a school of thought that says fans should ignore the monetary cost of acquiring any player. After all, its proponents say, it's the owners' money being spent, not theirs! A fan should care about winning, not team finances.
Is this a reasonable approach to take? Well, no. And a lot of the problem lies in the extremely narrow definition of cost that people consider.
Think of it like this:
A player's value to a franchise is tied to how much money he generates, which, for the most part, correlates with how many wins he produces. He is rewarded for this by being given X amount of money over Y years.
However, it's not all about money and years. There are other costs incurred in acquiring players, ranging from giving up draft picks to trading parts of the team. You could even make the argument that part of the cost of drafting Jeff Clement was not drafting Troy Tulowitzki.
Let's lump these into three main types of cost: money, years, and players. These affect two key aspects of any franchise - finances and the roster itself.
Firstly, finances. The money itself generally comes out of the owners. There's a slight cost to the fans incurred in ticket prices, etc, but that's not particularly important. The real blow comes when the payroll is considered. Loosely speaking, a team has a finite amount of money to spend on players each year. A very high salary for one player might not manifest itself in beer prices, but it certainly will impact a team's financial flexibility for as long as said player is on the books. Not only is Carlos Silva overpaid, but he's stopping us committing his slice of the payroll to a Ben Sheets or an A.J. Burnett. Conversely, this is why club controlled players are so useful - by getting production for $0, a club is free to pay more for improved production elsewhere.
Then comes the roster, which is slightly more subtle. The key point here is that there is a hard cap on how many players can be on an ML team, and then further caps on starting players. If a NL team had 28 MVP calibre players (who never got injured), 3 of them would be completely useless due to roster constraints, and a further 12 of them would be of marginal use, because they would spend the vast majority of their time on the bench or in the bullpen. A bad player on the bench might be free in terms of the money he makes but he takes up roster space that could be better used on someone else. Carlos Silva is one of the things preventing Ryan Rowland-Smith from starting in Seattle*. The idea of players being blocked due to superior play or clunky contracts is why you see very talented players get traded so often - a team with two 3-win catchers only needs one, after all. And of course, roster space is not the only part of the roster which can be considered a cost. Giving up a player means you lose out on the wins/money he generates over the life of the contract, which needs to be recouped somehow.
In other words you can't just trivialise a player's cost by saying 'it's not my money'. Acquiring a player affects a franchise in many different ways, and most of them have an impact on the quality of the team in the short and long term - which as a fan isn't something you can simply dismiss out of hand.
*A similar argument may be made demonstrating that a platoon is slightly less valuable than an equally productive single player.