In Academia, a soft money research position is one where uncertain money comes from an external source. The employee or xyr supervisor/boss might need to secure funding every couple of years or so, in order for the position to continue. The way out would be tenure or a government job.
An obvious soft money position is the post-doctoral fellowship, but many post-docs might not find a faculty position directly. For example, if the University of California at Los Angeles (UCLA) advertises a soft money assistant researcher position, they have a defined path of promotion steps. UCLA defines four steps of assistant researcher (2 years max), three steps of associate researcher (2 years max), and nine steps of researcher (3 years max until step IV, possibly indefinite from step V). If coming in at step I of assistant researcher, it might, theoretically, take 4*2+3*2+4*3=26 years of positions in steps of 2 years (first 14 years) and 3 years (last 12 years), until one might be appointed indefinately, if I'm reading things correctly. Needless to say, not ideal from an employee's point of view.
Although time-limited positions and soft money might not mean exactly the same thing, I suppose they often go hand in hand. That raises the question: how common is it for researchers to spend a long time, say more than 10 years, in soft money positions? It's one thing to "drop out" of Academia when finishing a PhD at 28 years, it's another thing to move from postdoc to assistant researcher to associate researcher, only to finally discover, at age 45, that you're not good enough for tenure, and lack the necessary skills and experience for a teaching position. Oops.
In Science, Technology and Engineering (STEM) fields in the USA and Canada, how common is it for researchers to spend more than 10 years in time-limited soft-money research positions?