The presumption in favor of judicial review of agency action is a cornerstone of administrative law, accepted by courts and commentators alike as both legally appropriate and obviously desirable. Yet the presumption is puzzling. As with any canon of statutory construction that serves a substantive end, it should find a source in history, positive law, or the Constitution. None of these, however, offers a plausible justification for the presumption. As for history, the sort of judicial review that the presumption favors — appellate-style arbitrariness review — was not only unheard of prior to the twentieth century, but was commonly thought to be unconstitutional. The ostensible statutory source for the presumption — the Administrative Procedure Act — nowhere instructs courts to strain to read statutes to avoid the preclusion of judicial review. And although the text and structure of the Constitution may prohibit Congress from precluding review of constitutional claims, a presumption responsive to constitutional concerns would favor review of such claims, not any and all claims of agency wrongdoing.Bagley's focus is on American law, but the question of the basis for judicial review of administrative action is one which preoccupies judges and jurists alike in many jurisdictions. English academics have spent two decades with horns locked over the question.
To date, however, the presumption has gone unchallenged. This is regrettable. Congress has the constitutional authority, democratic legitimacy, and institutional capacity to make fact-intensive and value-laden judgments of how best to weigh the desire to afford private relief against the disruption to the smooth administration of public programs that such relief may entail. Courts do not. When the courts invoke the presumption to contort statutes that rather clearly preclude review to nonetheless permit it, they dishonor Congress’s choices and limit its ability to tailor administrative and regulatory schemes to their particular contexts. The courts should end this practice. Where the best construction of a statute indicates that Congress meant to preclude judicial review, the courts should no longer insist that their participation is nonetheless indispensable.
For my part, I would state the presumption of reviewability slightly differently: where there is an effect on an individual's rights or interests (and the individual can scale procedural hurdles, such as standing), administrative action is presumed to be reviewable. This, I think, is the Canadian position and the position in other Commonwealth countries. So stated, the presumption is simply an aspect of the rule of law.
One argument that Bagley does not consider is the one made most elegantly by David Dyzenhaus. Where a statute purports to exclude judicial review, it speaks out of both sides of its mouth: it imposes limits yet asks courts to ignore them. Here is what I said about the problem in A Theory of Deference in Administrative Law:
It might be argued that in order to truly give effect to legislative intent, a reviewing court would have to give full effect to a statutory provision that purports to exclude judicial oversight. This is not necessarily so. The problem with privative clauses, so called because they deprive a reviewing court of jurisdiction, is that they create tension within a statute. On the one hand are provisions that prescribe limits to statutory powers; on the other hand is a privative clause which purports to render these limits unenforceable by reviewing courts. Privative clauses are, as a result, incoherent, as the much-maligned Lord Hewart observed in the context of a “conclusive evidence” provision:I also had something to say about the Administrative Procedure Act, suggesting that the best way to resolve the tension created by privative clauses is to ratchet up judicial deference, rather than excluding judicial review altogether:
In passing such a clause Parliament, it may be thought, was really stultifying itself, because, having inserted express provisions in the Act for the protection of persons liable to have their property taken without their consent, and having enacted that the council in making…an order must have regard to these provisions, it then, by means of this ‘conclusive evidence clause’ rendered such provisions nugatory, and, so far as victims are concerned, a mockery.
Section 701(a) of the APA provides that judicial review will be generally available ‘except to the extent that (1) statutes preclude judicial review; or (2) agency action is committed to agency discretion by law’. The presence of these clauses gives rise to the same incoherence that is created by the presence of a privative clause in a statute. Indeed, the situation created by the APA is indistinguishable from the situation created by an individual privative clause, for the APA only legitimates the exclusion of review; that exclusion is, in the final analysis, effected by privative language in the statute. Where language in the relevant statutory provisions satisfies the two clauses in the APA, it has a privative effect. Such language is permitted, but not mandated, by the APA. The incoherence could be removed by treating the satisfaction of the two clauses as factors going to the choice of the appropriate standard of review.Bagley notes that intrusive judicial review can undermine statutory schemes: he gives the example of the Veterans' Administration. I agree, but I think the solution lies in deference to agencies rather than in the exclusion of judicial review.
You can download the thought-provoking paper here.