The majority of fantasy baseball leagues are of the one-year variety, but a significant number allow you to keep players for multiple seasons. A typical structure in a league with a $260 budget allows you to keep a player at his purchase price for up to three years, and if you want to extend him beyond that, you'll have to add $5 per year.
For example, in the RotoWire Staff League, Cliff Lee was in his third year with "Liss and Shoe's Dynasty" at $5 (it was actually Lee's fourth but because we got him as a reserve in 2008 rather than in the auction, his "service time" clock started in 2009 for our purposes), and we extended him through the end of next season at $10 per season. That means he'll cost us $10 in 2011 and another $10 in 2012. We could have simply paid $5 this year and let him go as a free agent after 2011, or we could have paid $15 per year through 2013, $20 through 2014, etc. (More on why I only extended him one extra year below).
The effect of our team having Lee at $10 this year rather than the $25 - $30 he'd be worth in a normal auction means I have $15 - $20 more dollars to spend on the remaining player pool than I should. Given that each of the 18 teams can keep up to 15 players, plus minor leaguers who might have gone for auction dollars, you can see how much extra money is available for the unkept players. Of course, Lee at $10 isn't typical - it's a better than average bargain, so let's say the average discount on a kept player is more like $8 rather than $15 to $20. And let's further assume that the average team keeps 12 players. That means there are 18 teams times 12 players times $8 which comes out to $1728 extra dollars to spend on the 198 roster slots (11 remaining slots times 18 teams) that need to be filled. That averages out to roughly $9 extra per player.
But this extra $9 won't be applied uniformly but in proportion to a player's value above replacement. In other words, there will still be $1 players who are barely better than those who go unpurchased, and the majority of the inflation will skew toward the top. Moreover, the scarcity of unkept star players and the presence of teams with many cheap keepers and nearly full budgets to spend on only half the roster slots will also drive up prices. For example, Albert Pujols went for $70 (as opposed to his usual $45-$50 in a deep mixed league), Miguel Cabrera went for $59 and Alex Rodriguez went for $56.
While the exact degree of inflation will depend upon the number of players kept, the rules for keeping and extending players and the parameters of the league, it's almost certainly going to be a significant factor in determining player prices. As such, you'll need to factor it in when deciding whether to keep players or throw them back into the pool. For example, we kept a $28 Mariano Rivera and a $29 Derek Jeter, players we probably would have tossed back were it not for expected inflation.
Finally, one note on why we extended Cliff Lee only through 2012 at $10 rather than through 2013 at $15 or 2014 at $20. For starters, pitchers (even ones as reliable and talented as Lee) often get hurt or lose effectiveness. Consider Johan Santana, Brandon Webb or Jake Peavy, all Cy Young award winners just a few years ago. Moreover, getting Lee at $10 for the next two years is better than having him at $15 for the next two years, i.e., we maximize our chances of winning in the near term while we still are fairly sure Lee will be good. Lastly, there's a time-value of money consideration - present value is always worth more than equal future value. The counterargument is that in this inflation-driven league, Lee would go for $45 on the open market (C.C. Sabathia's 2011 price). In that case, we profit $35 a year for two years or $70 total. If we went $15 per at three years, that would be $90 total. At $20 for four, it's $100. But those numbers need to be discounted for both the time-value of money and also the decreasing certainty with which we can project players as we look farther into the future.