Yes, great graph.
It illustrates the fundamental truth overlooked by most retirement planning schemes: the stock market doesn't just automatically grow in value over time. It grows in emphatically punctuated booms driven by technological advances.
The boom of the 1920s owed a lot to the telephone and automobile, linking businesses together in new ways. The boom of the 1960s was mainframe computerization, and the boom of the 1990s was personal computers and Internet connectivity. All these permitted entrepreneurs and established businesses to gain ever larger leveraged multipliers of turning effort into impact and value and wealth.
No breakthrough technology means no boom. We won't have another until another such technology arises. The Internet has largely plateaued in terms of business value. We don't have anything obvious on the horizon that will create a 10x productivity multiplier over email or Excel or StackOverflow, in the way that computers replaced adding machines and email replaced snail mail. Judging by the history of industrialized society, something will arise eventually to spark another boom (my best bet is neural-computer integration), but we can't say what or when. The next 20-year quadrupling of the stock market may begin in 2015 or in 2060.