Philip Morris is remarkable on a number of levels it seems to me, the most immediate of which is how the Court divided: what issue would you have predicted that Breyer, Roberts, Kennedy, Souter and Alito would line up against Stevens, Thomas, Ginsburg and Scalia?
On a somewhat more substantive level, it seems remarkable for how quickly the Court is stealing the "tort reform" thunder from the Republican Party.
On a clearly more substantive level, this decision continues the growing "jurisprudence" that has accreted, in about a decade (since BMW v. Gore (1996)), establishing firm constitutional limits on the "jury system" across the states. That jurisprudence is now quite substantial: TXO Production v. Alliance Resources (1993) (awards that are "grossly excessive" can violate fundamental fairness and therefore DP); Honda Motor v. Oberg (1994) (punitive damages awards by juries must be subject to judicial review); BMW v. Gore (awards can be grossly excessive if they bear no reasonable relationship to the damages caused by the tortfeasor, if the tortfeasor's conduct isn't sufficiently "reprehensible" or if the award is excessive compared to like cases); Cooper Industries v. Leatherman Tool (2001) (the judicial review of jury verdicts awarding punitives must be "de novo"); State Farm v. Campbell (2003) (an award where the ratio of punitives to consequential damages award was 145-to-1 is presumptively excessive). . . .
What has been of deepest interest from a constitutional perspective in this line of cases, imho, is the evolving justification for limiting punitive damges. What the Court has said, as far as I can tell, is that because civil defendants are not accorded the protections afforded criminal defendants, punitive damages pose an acute danger of arbitrary deprivation of property, which is heightened when the decisionmaker is presented with evidence having little bearing on the amount that should be awarded. Thus, in Philip Morris where some of the instruction may have been interpreted as directing the jury to base the award on harm done to other victims (not the one then in court), this majority seemed hell bent on protecting Philip Morris's "property" from the potentially arbitrary deprivation an excessive damages award would represent. This is both a curious definition of "property" and a curious use of the Supreme Court's increasingly scarce capacity to resolve cases.
Of course, complicating this "code" is the fact that the head counts in a couple of these decisions (and some others not mentioned) muddled the results issued and that the justifications for these holdings have, when characterized in subsequent opinions, well, drifted substantially. As for Philip Morris, my colleague Bill Childs at TortsProf puts it this way: the decision seems "prone to causing confusion and difficult to imagine trial courts finding helpful."