While I agree that it is easy to spend lots of money and get no schedule improvements, it is possible to spend money in ways that will reduce schedule.

For instance if you need a Perl project done yesterday and money is no object, I can name [link|http://www.stonehenge.com/|consulting companies] that can do it.

More generally, most projects have some parts that could be parallelized if you had additional resources. Usually companies have those resources, but they are not available because they are doing other things. If you're willing to pay the cost of having those other things not happen, you can add temporary resources.

And as you say, spending up front is usually the best way to achieve schedule reduction.

However the tradeoffs that you can achieve are limited. And you quickly find that the schedule reduction that is achievable is limited, while the cost of those diminishing returns raises quickly. That's what I understood the observation that "schedule is the least flexible" to mean.

Furthermore most people who control these budgets don't know enough about IT to understand what ways they can spend money and save on schedule, and what ways they can spend it and blow their schedule. Given how greatly the latter outnumbers the former, their odds of achieving a schedule reduction through spending money are minimal.

Cheers,
Ben