Knowledge retention, really?

The topic of knowledge retention and the aging workforce came up in my KM class this week.  It is generally acknowledged to be a problem to have long-experienced people retiring or otherwise leaving companies.   The class asked the sensible question of "how real is it?"  In conjunction with this, Dorothy Leonard (of the wonderful Wellsprings of Knowledge) has an article on the issues of knowledge retention in the May 1, 2005 CIO Magazine in conjunction with her new book with Walter Swap, Deep Smarts: How to Cultivate and Transfer Enduring Business Wisdom.

CIO, May 1, 2005, Dorothy Leonard, How To Salvage Your Company's Deep Smarts:

The approaching exodus of retiring baby boomers will severely erode the knowledge base of many companies. Fortunately, there are ways to re-create this crucial expertise.

... But people with deep smarts can be indispensable. Why? Because their particular brand of expertise is based on long, hard-won experience. They are the go-to people known for their swift, seemingly intuitive judgments. Such experts differ from their less experienced colleagues in having the ability to view a problem at a system level and yet dive into the details when necessary—and identify a familiar pattern.

There are a slew of good comments to the article as well, including some that do not completely agree with the sentiments expressed.  To that end, a couple students questioned the premise of knowledge retention - that long-term employees are retiring in great numbers, creating huge risk for companies.  Some counter-arguments or questions that test the premise:

• Is it real?  Has the cost of this retirement been documented, or is the risk all anecdotal?
• What is the real distribution of employees in the at-risk companies?  (Distributions in terms of age, longevity and nearness to departure)
• How many people stay at one company for more than five years to accumulate "deep smarts" about the company?  We are told that the current workforce will be in many companies and careers through their lives.  Thought of this way, it really sounds like these programs are "brain dumps" that the experts will resist with all their might.
• How much can truly be "salvaged" this way?  (The Leonard article, and I assume the book, talk about establishing processes that ensure training of less experienced people to help them understand the thinking of the experts.)