Bob, I was hoping that "putting aside" would divert just the sort of comments you started with. However your last paragraph offered an answer to the question.
Too simple though.
Re discussion here of public preferences and willingness to pay, without referenda on every proposal, it's evident that the public doesn't get to choose among particular policies. They vote for parties or particular candidates to represent general preferences. In elections, voters approve or reject the overall acceptability of the program -- the portfolio of policies -- a party/candidate represents.
If as the column suggests the portfolio of the incumbent regime imposes overall more costs than benefits, I suggest the evaluation of any particular policy should reflect that. In practice, that is the way the electorate is going to evaluate it -- especially in parliamentary governments.
Related to that, the viability of any carbon tax scheme depends on its continuity. As noted by Jesse and others here, Australia's carbon tax was finished when the party that imposed it was rejected as a whole and replaced.
The column also suggested a spillover effect of policies across jurisdictions. The author suggests that the viability of one policy adopted in BC was negatively affected by policies in California to which it was connected.
There's an analogy in healthcare: A patient with multiple conditions often has to take several medications, each prescribed by a different doctor for a particular illness. Medications often interact in ways that may block effectivenss of some or even have toxic effects. Only recently has an increased effort been made to coordinate care to make sure that all the treatments the patient receives work together to heal rather than to harm. We don't congratulate the doctor for prescribing the pill that cured the patient's skin rash but that, in combination with another medication, destroyed his liver.