It's been awhile, but I've finally gotten unburied long enough to begin to catch up on my Stavins' columns. Here is the meat from his dozenth "An Economic Perspective" piece in the March/April 2006 The Environmental Forum (A Utility Safety Valve for Cutting CO2):
The safety valve would address cost uncertainty through a simple mechanism. After allocating emissions permits freely to power plants, the state governments would announce that they will sell — note, sell, not give away — additional permits at a fixed price. That price instantly becomes a cap on compliance costs, and eliminates the cost uncertainty that otherwise plagues the program. ...
Importantly, this mechanism is only triggered if costs are unexpectedly high, whereby the safety valve offers important economic protection, while still providing powerful incentives for emissions reductions. On the other hand, if environmental advocates are right, and compliance costs are low, the safety valve would not be activated.
If you remember, Massachusetts wanted the safety value. The bad news is that the other RGGI states didn't:
The safety-valve plan for RGGI could have been good news for the environment and the economy, and a good model for an eventual national program. As it turned out, however, there seemed to be more interest from some states in short-term symbolic actions that in real long-term achievements, and the RGGI agreement went forward without the safety valve and without Massachusetts.