Justifying judicial review in division of powers cases

J.S. Mill famously argued that where federal governance divides final decision-making powers between two or more levels of government, the further "power to decide between … (the different levels) in any case of dispute should not reside in either of the governments, or in any functionary subject to it, but in an umpire independent of both." A.C. Dicey suggested that this requirement provided a decisive reason not to adopt federal forms of governance: federalism requires judicial review, and judicial review undermines the parliamentary sovereignty that is constitutive of just governance. Yet subsequent federal theory and practice place pressure on any such entailment claims. Wheare (1946/1953: 64-66), Duchacek (1970: 256ff), and Elazar (1987: 84) all note non-judicial means of reviewing governmental conduct and disputes, notably including legislative second chambers and referenda. States like Ethiopia and Switzerland rely on such non-judicial bodies for some reviews.¹ Later debates about whether courts are well-placed to make decisions about inherently normative phenomena, like rights, may also implicate the case for judicial review of structural questions, like those concerning federal divisions of powers. Adrienne Stone, for example, argues that questions about the division of powers are, like rights, irredeemably normative and courts lack special expertise on normative queries. One may plausibly contend that legislatures have more legitimate claims to authoritatively interpret contested concepts.

In a forthcoming piece in the Journal of Legal Philosophy, "Judicial Review and the Basic Architecture of Federalism", I return to the more functionalist approach implicit in Mill and Dicey to defend judicial review on non-rights-based terms. I argue that a third-party arbiter is not only necessary to resolve disputes between different levels of government, as Mill noted, but also to police the constitutionally defined borders of each level’s jurisdiction.² Developments above establish that accepting federalism need not entail accepting judicial review. However, some features of well-designed courts, combined with issues with other forms of review, provide a defeasible case for judicial review in division of powers cases. This case stands regardless of how one views rights cases. Unlike Dicey, I do not take it as a reason to reject federalism.³ It is, instead, a reason to ensure courts in federal states are designed to serve both necessary roles.

The need for a third-party umpire is familiar – no one should be the judge of their own case – but the need for third-party border police may be less so. The basic idea is that the need to ensure different levels of government stay within their own constitutionally defined powers does not only arise in the context of live disputes between them. For example, if provinces choose to let a federal government pass a law in an area within provincial jurisdiction, that does not change the fact that the law should be considered ultra vires the federal government. For another example, even if one desires cooperation between levels of government, someone must ensure that each cooperating party is acting within its own constitutional mandate. This is necessary not only to ensure constitutional supremacy but also to further the desired ends of federalism. If, for example, politically weak provincial governments agree to cede authority to federal governments, they are unlikely to serve as bulwarks against undue centralization. While federal, provincial, etc. governments could, in theory, police each other, structural incentives highlight the need for a third party. A rational government actor will seek to interpret powers in ways that expand the scope of their own authority – so they can pass laws they want – and to contract powers to leave responsibility for more difficult, politically costly decisions to other levels of government.

The need for a dispute resolution and border policing body does not, of course, entail the need for judicial dispute resolution or border policing. But properly designed courts should be well-suited and incentivized to serve both roles better than alternatives. Dispute resolution and border policing are traditional judicial functions. Even Dicey admitted that courts should be constitutively committed to protecting constitutions. Suitably independent courts will not face incentives to deviate from the constitution to protect others. Indeed, constitutional fidelity should be their primary interest when deciding cases. Institutional constraints on courts in both common and civil law jurisdictions further limit courts’ ability to stray from constitutional norms. The law of precedent and civil law textual fidelity limit deviations from constitutions.

These considerations provide the core of a basic case for judicial review in separation of powers cases. Courts’ stability-conferring role and expertise are also sometimes raised to support judicial review. I suspect each is too contentious to establish a strong case for judicial review on their own. However, each usefully supplements my own, primarily functional, case. Likewise, while I believe the case for judicial review in division of powers case need not rely on rights, I find fellow former Federal Scholar in Residence Erin Delaney’s rights-based argument for federal judicial review compelling. Courts in federal states may also protect rights.

This case for judicial review is, again, defeasible. Courts are clearly imperfect. I suspect any reader of this blog can easily identify a case where courts in their jurisdiction failed to serve either role above well. I spend considerable time in the article addressing empirical challenges to my arguments. The defeasible case nonetheless stands absent evidence that another entity can play necessary dispute resolution and border policing roles. I alluded to incentive-based reasons to question whether legislative or executive actors can play those roles above. Such incentives arguably need not apply in referendum or specialist second chamber cases. Yet existing evidence does not suggest that either offers a compelling alternative to judicial review. Regular referenda risk instability – and likely alter the constitutional division of powers, rather than policing it. It is thus unsurprising that even Switzerland, which is often offered as the core exemplar of referenda-based accountability, permits some judicial review in relevant cases. The main example of a specialist second chamber, Ethiopia’s House of the Federation, is, in turn, often critiqued for, among other issues, failing to properly resolve disputes or police borders.⁴

The case for judicial review in separation of powers cases thus stands. Does that simply vindicate the status quo? There are at least two reasons to think not. First, note that the case at hand relies on well-functioning independent courts committed to the constitution. Courts can only serve their intended role under certain conditions. Those conditions do not obtain universally, calling for reform in some jurisdictions. Second, if the case for judicial review rests primarily on the protection of the constitution itself, this may call for differences in how courts approach cases. Courts should, for example, ensure that they clearly and consistently delineate distinct domains of federal, provincial, etc. jurisdiction to avoid border collapses. And many courts should be less deferential to other levels of government in how to interpret powers. These challenges to the status quo are just two of the issues introduced in the article that I lack space to address in detail. I hope that further work will scrutinize my case for judicial review in division of powers cases and its possible implications for these more practical concerns.

As the 2024 Federal Scholar in Residence at the Eurac Research Institute for Comparative Federalism, I had an opportunity to workshop an earlier draft of the text with Institute staff and a visiting researcher. The article no doubt benefited from this direct engagement with my work and other opportunities to discuss federal practice and theory with Institute staff and participants in the Winter School on Federalism and Governance. I look forward to future collaborations with members of the Institute and the wider community that it helps foster.