Can super-agents raise player salaries?

Robert, a loyal MR reader, asks:

I was recently
reading about ARod’s decision to leave the Yankees. The article
mentioned "superagent Scott Boras." It’s widely believed in the sports
community that Boras has the ability to increase the salaries beyond
what they would get with a regular agent. Considering that there are
only 30-odd teams that might want a player, I find it hard to believe
that an agent could make such a big difference.

I know more about Mark Alarie than ARod, but super-agents may matter through the following mechanisms:

1. The super-agent manages an otherwise incompetent or unruly player.  The agent is about improving the quality of the player as much as extracting surplus from the team.

2. A super-agent, especially if he has repeat business with teams, may credibly certify the unobservable qualities of players, even star players.

3. Boras may be very good at marketing his players to management and getting owners to open up their pocketbooks.

4. If Boras represents multiple stars, clubs will be reluctant to cross him.  The equilibrium here is tricky.  But if the agent has discretionary power to steer a player to one equal offer or the other, and the club reaps surplus from each player, a club may overbid for one player to stay on the agent’s good side and receive favorable discretionary treatment later on.  Repeat dealings with the agent also mean that the club is more likely to follow through on its implicit commitments to both the player and the agent.

5. Robert suggests that some (non-super) agents may in fact be in league with the owners, not the players.

Can you think of other possible mechanisms?

Addendum: Here is J.C. Bradbury on Boras.  And here is a New Yorker story on Boras, courtesy of John de Palma.