He provides a game-theoretic description of judicial discretion.
Imagine a unicameral Parliamentary system with no particular transaction costs in producing legislation. The executive and the Parliamentary majority have a unified ideal point.
If the composition of Parliament and the Executive have changed since legislation was passed, and a case comes up revealing potential ambiguity in interpretation, the Judiciary can choose to interpret consistently with the bargain that was struck when the legislation was passed, the outcome that might obtain if the legislature and executive were to reconsider it, or the Judiciary's own view of what the public interest requires.
The judiciary has zero discretion in that case. If it returns a decision inconsistent with Parliament's intention or views, Parliament immediately legislates to correct.
In a bicameral system, legislating to correct a judicial decision requires agreement between the two houses. The executive will sit in the House; it needs the agreement of the Senate. If views between the two houses differ, the judiciary has discretion within the Pareto set: the set of all points between the ideal point of the House and the ideal point of the Senate. If the judiciary sets a decision outside of that Pareto set, the legislature reverts to some point within the Pareto set.
He illustrates as follows.
He illustrates as follows.
In a unicameral Parliamentary system in a zero transaction-cost world, there is no opportunity to diverge.
In a unicameral Parliamentary system that has a coalition, there is opportunity to diverge if the governing coalition does not understand the game or refuses to play consistently with it.
If the bargain within the coalition is weak, the judiciary's discretion is constrained to the Pareto set of the members of the coalition. If it produces a decision outside of that range, the coalition can negotiate to overturn, returning legislation to the Pareto set. But it will not be able to find agreement to legislate to overturn a decision within the Pareto set: by definition, at least one member will prefer the Judiciary's stated position to the status quo.
A more clever coalition will realise that this game gives the judiciary room to unwind the bargain struck during coalition negotiations, and will pre-commit to overturn any decision that starts down that path - even if one party prefers the judiciary's decision in that particular case.
That's the zero transaction cost world. A not-stupid coalition precommits to not letting the judiciary play shenanigans. And so the judiciary does not engage in adventures.
Now let's move to the more realistic positive transactions cost world.
Parliaments have a habit of producing bad legislation whether through haste, incompetence, or unwillingness to resolve political conflicts within a governing coalition. In that latter case, explicitly political decisions may have been avoided through use of ambiguous language that will require the judiciary to take interpretive decisions of political consequence. You could imagine the legislation as not providing a point on the line, but rather a fuzzy shaded area spanning potential interpretations. And you'd hope that the legislature at least would have set legislation ruling out interpretations outside of the Pareto set.
Legislating around a decision is not costless. A governing coalition has its own legislative priorities. Time, effort, and drafting resource spent bringing an errant decision back in line means time, effort, and drafting resource not spent on other pieces of legislation. And some MPs' understandings of comity give the judiciary much room for shenanigans before the legislature would be allowed to act.
These transactions costs widen the range of judicial discretion. Even a governing coalition with tight agreement will not act unless the judiciary strays beyond a tolerable range. Beyond that range, the legislature will correct aberrant decisions. Within the range, the judiciary has discretion.
And that turns things into an expectations game. If the judiciary expects that Parliament faces high transactions costs for reversion, then it will play as though it has a very wide range for discretion. Repeated refusals by the legislature to correct aberrant decisions affect those expectations. They reinforce the judiciary's view of its own discretion, and solidify legal academics' views that the court actually has that discretion.
Smith v Fonterra looked like judicial shenanigans. Emissions have cumulative effects on global warming. No individual emission is the problem. It's their cumulative effect. Regulation makes far more sense than approaching it as tort. And we have a regulatory system around emissions. In the case of energy sector defendants, their emissions are fully covered by the ETS.
It took far too long to do it, but Parliament has finally moved to correct.
I doubt it will be enough to convince the Supreme Court that the legislature is generally willing to incur costs to correct adventures by the judiciary. But it is a very good start.
