Subsidiarity and the Allocation of Governmental Powers

Abstract Every country must allocate final decision-making authority over different issues/subjects within its boundaries. Historically, many scholars working on this topic implicitly assumed that identifying the features providing entities with justified claims for authority and the entities possessing those features would also identify which groups should have which powers (or vice versa). However, many candidate allocative principles select multiple entities as candidates for some sub-state authority and yet fail to explain which powers each should possess. Further work must explain which groups should possess which powers when and what to do when two groups can make equally-valid authority claims using the same principle. Subsidiarity, the principle under which authority should presumptively belong to the entity representing those ‘most affected’ by its exercise and capable of addressing underlying problems, is one of the few principles focused on identifying which groups should have which powers. Unfortunately, subsidiarity alone does not provide guidance on many issues/subjects. Useful subsidiarity-related guidance relies on balancing underlying justificatory interests, which do the real allocative work. Another allocative principle remains necessary. A deflationary account of subsidiarity’s allocative potential nonetheless provides insights into how to articulate a new principle and accounts of subsidiarity that can fulfill other moral roles.

preferable. 4 If subsidiarity-based allocations are then consistently plausible, subsidiarity is valuable: it resolves a common, persistently challenging problem. Exploring competing conceptions' allocative potential could help us choose between them or highlight the need for other ways to establish subsidiarity's value.
My analysis ultimately demonstrates that subsidiarity as such is not a useful allocative principle. Subsidiarity does not provide guidance on many issues/subjects, especially regarding divisions of constitutional powers. Useful subsidiarityrelated guidance balances underlying justificatory interests, which do the real allocative work. At least one other allocative principle is accordingly necessary, and proponents must justify subsidiarity's value in other ways.
To establish this conclusion, I first outline the need for allocative principles and how conceptions of subsidiarity purport to allocate authority within countries. I then argue that no existing conceptions of subsidiarity fulfill the desiderata of a plausible allocative principle-namely, the need to provide concrete, intuitively acceptable guidance on which entities should possess authority over particular subjects (and when) that can at least apply in liberal democracies and to explain at least a de minimus number of real-world allocations of final decisionmaking powers-and subsidiarity's characteristic feature, a presumptive allocation of authority to the 'closest' level of governance. I further argue that any account of subsidiarity that appears plausibly capable of providing concrete guidance collapses into a commitment to supporting principles that make the further appeal to subsidiarity redundant-and could lead to a misallocation of authority where supporting principles do not actually support 'close' rule. Despite this deflationary conclusion, analyzing subsidiarity's allocative potential remains valuable. It explains why and when the principle seems compelling (and, consequently, why one may wish to identify another moral role it could fulfill) and it highlights the need for and desiderata of another allocative principle, the pitfalls candidate principles may face, and a method for identifying a better principle.

The Importance of Scrutinizing Potential Allocative Principles
Contrary to present scholarly trends, 5 questions about how to allocate authority within countries should not be confined to federal and provincial (including canton, länder, US states, etc.) governments, federations, or constitutional moments.  A compelling allocative principle should guide decisions about which powers should rest with which government in federations, but this need not apply to federal and provincial governments alone. For instance, cities may also be relevant. 6 Moreover, countries need principles for when authority should rest with which groups for their everyday functioning. Central governments require principles for which powers they should delegate to others and when. A plausible authority allocation principle should provide guidance on the subjects over which entities should have authority in constitutional divisions of powers, where authority over issues should rest, and, by extension, when one should delegate it. This principle should allocate authority over subjects and issues in centralized and federal countries. Recent developments challenge traditional allocations and allocative principles in ways that highlight the importance of-and require further work onauthority allocation. Rawlsian political philosophy initially assumed unitary countries and only gradually explored how minority rights and federalism impact the structure of just societies. 7 Theoretical discussions of authority allocation thus often occurred in law and political science; focus on constitutional divisions of power was common. 8 While some works addressed basic allocative concerns, common emphases on narrower issues (e.g., areas of 'overlapping' federal and provincial jurisdiction) was understandable given the low chances of constitutional change and generation of plausible allocative principles that explained real, prima facie justified decisions. 9 Yet recent events challenge the allocative status quo. For example, sanctuary city debates and demographic differences between municipal and rural residents led to calls for greater authority for cities. 10 For another, cities, Indigenous nations, and other sub-state 'national' groups demanded increased authority over public health management after purportedly unresponsive or otherwise problematic central/federal/provincial actions related to  At the same time, political philosophers increasingly recognize that principles 'justifying' provincial control over some topics equally 'justify' control by other entities. Per Daniel Weinstock, such principles (analyzed below) equally, if not better, justify municipal authority. 12 So, countries that "incorporate : : : [decentralizing principles] should on pain of arbitrariness apply them to cities." 13 Avner de-Shalit, Ran Hirschl, and others make similar claims; per Hirschl, the principles also justify 'rural' authority. 14 Helder De Schutter et al. then argue that even sub-state and/or sub-state national claims. For example, Heather Gerken stresses the importance of addressing a broader array of claims to, and seeming exercises of, authority. See Heather Gerken, "Dissenting by Deciding" (2005)  'nations' who lack their territory still possess unique cultures that could justify forms of non-territorial autonomy (and thus sub-state authority) on the selfdetermination-based grounds justifying provincial control in some countries. 15 Accounts of authority (and political justice at large) should explain if, why, and when these demands should be fulfilled. These developments also highlight an often-overlooked distinction between features providing entities with justified claims for authority and factors specifying which groups should have which powers. For instance, self-determination is often discussed as if it answers the ontological question, 'What entities can justifiably possess powers in a country?', which addresses the features that make groups valuable in ways that could justify providing them with some authority, and the more specific allocative question, 'Which entity should possess a specific decision-making power?' 16 Yet if, for example, a province and a city each have self-determination-based claims to control language policy, another principle must allocate authority between them. Several scholars view parallel arguments for provincial and municipal control as evidence that those who allocate powers to provinces should allocate them to cities, too. 17 Yet they might just make both of them the kinds of things that could have some authority. Other considerations should explain which powers each should possess and when. They may not require more constitutional powers for cities, but a complete account of authority allocation should address municipal claims.
We must accordingly separate the ontological and allocative questions and scrutinize the genuinely allocative potential of candidate principles for resolving the latter. Detailing issues with existing principles clarifies the problem. To begin, most existing principles fail to specify the precise powers entities should possess. For instance, the scope of moral self-determination rights is contested. Claims that they should entail powers over cultural policy for unique cultural groups are plausible. 18 Yet institutional forms other than sub-state autonomy may best protect cultures. 19 Moreover, even a 'right' to authority (in the sense at issue) could be defeasible. Self-determination rights can be constrained by, for example, the need for broader state stability. 20 Relationships between features grounding self-determination rights and particular issues/subjects are also unclear, leaving self-determination underspecified as allocative guidance. For instance, while some claim that self-determination rights-holders should control health policy, 21 culture and health are hard to link absent controversial posits about health's cultural determinateness, and considerations above seemingly support central control. 22 International law's minimalist commitments on the scope and institutional forms of cultural autonomy are thus unsurprising. 23 Other candidate allocative principles also raise problems: 'decentralizing' one's appeal to entities' unique interests/values or challenges; democratic benefits of providing authority to the entity 'closest' to the citizens capable of addressing it or best reflecting the interests of the 'most affected'; or epistemic benefits of 'local' control. 24 Yet provinces, cities, and sub-state nations could receive some authority under each. No principle clearly specifies which entity should possess which powers where multiple entities possess equal claims under that principle. Attempts to break justificatory 'ties' then often beg questions. Groups may, for instance, disagree on how to draw the boundaries of 'most affected' groups or importance of territory for protecting cultures. 25 'Most affected' status or geography cannot then be justificatory tiebreakers.
The prima facie justification of various authority allocations further suggests that the ontological features of candidate authorities cannot determine all allocative questions. There is, for instance, wide variance on where healthcare authority resides in federations: self-determination-based decentralization decisions and coordination-based centralization decisions each seem legitimate in some contexts. 26 At best, one must scrutinize the underlying principles to explain why prima facie justified choices are not fully justified. 27 The features of entities with prima facie justified authority claims alone cannot resolve these issues. A theory is likely better if it explains how the features that make an entity ontologically valuable could justify allocating powers to it, but different principles may apply at each stage. While self-determination and subsidiarity are discussed as competing principles, 28 subsidiarity could guide allocating powers amongst entities with equally compelling self-determination 'rights'-based claims. A province and a city could both have extraordinarily strong self-determination rights with an epistemic understanding of subsidiarity best distinguishing their claims to particular powers.
A genuine allocative principle capable of addressing the preceding should provide guidance on which powers should belong to which entities in federations where powers are formally separated, what federations should do in areas of shared jurisdiction, when authorities in all countries should devolve to other entities, and where authority should rest where no entity possesses explicit powers. Ideally, it should also specify which powers should belong to which entities in a way that secures stability, but it should at least explain when decisions should rest with each entity. A fully action-guiding principle should explain why and when particular entities should possess decision-making powers. To play an oft-intended "democratic structuring" role, 29 it should explain which entities should have authority over which subjects, not merely when it should be able to decide an issue. This need not require that all parties to a debate share one understanding of a principle or its implications. It only requires that the principle have some basic features that present distinct moral reasons favouring particular allocations. For instance, while self-determination admits several interpretations and its application in real cases will often be contested, self-determination-based claims are usually grounded in moral reasons and particular groups should possess particular powers for common ends. 30

90
Da Silva appeals to, for example, the epistemic value of particular allocations even absent perfect specifications of when and how the related arguments differ. 31 It is possible, even probable, that no one principle plausibly answers all allocative questions. Even then, one should explain the restricted scope of any principle one invokes. For instance, if different considerations apply to devolution and federal divisions of power, one should explain why this is so and why narrowly-tailored principles applying only to one are desirable. If, in other words, asking one principle to answer all allocative questions is unfair, we should still examine which candidates offer distinct, intuitively compelling results such that they should be operative principles stakeholders can validly raise to support particular claims in a context. I will now demonstrate that appeals to subsidiarity do not establish a general presumption that the most local entity should possess authority as claims, and the presumption's seeming plausibility in some cases is best explained by other (e.g., democracy-or epistemic authority-based) principles. Moreover, real-world applications do not provide intuitively compelling results favouring local control. Subsidiarity, then, should not be even one of our standalone allocative principles. Any value that the concept may have will stem from other potential moral roles.

Subsidiarity's Allocative Potential
Subsidiarity, again, holds that authority should presumptively belong to the entity representing those 'most affected' by its exercise and capable of addressing underlying problems. Subsidiarity, too, admits multiple interpretations. 32 Yet this provisional understanding refers to an influential concept. The EU's power-sharing rules reflects its basic features. 33 Subsidiarity is also oft-discussed as a means of structuring divisions of governmental powers at international and domestic levels. 34 While it could fulfill other functions 35 -and the best theory thereof may answer questions about, for example, how to allocate powers between state and non-state entities, rather than allocating constitutional powers within states or state-like entities like the EU 36 -subsidiarity is often discussed as the best principle for allocating authority within countries. 37 Its application to this task is often taken as self-evident: its "most obvious implications are for constitutional ordering within" countries. 38 Subsidiarity promises to allocate authority over issues and subjects. 39 While it is viewed as most compelling for dividing powers during constitutional transition, 40 its allocative potential within countries is supposed to extend to any allocative moment, constitutional or otherwise, such that it identifies when it is appropriate to delegate constitutional authority and create quasi-constitutional forms of revocable authority. 41 Subsidiarity is thus a plausible starting point for addressing authority allocation questions. Addressing subsidiarity's allocative potential should also contribute to subsidiarity studies: if subsidiarity cannot fulfill this role, accounts should attend to its other possible roles (e.g., its ability to plausibly resolve state v. nonstate conflicts). 42 Cataloguing subsidiarity's forms and purported justifications suggests that subsidiarity could guide authority allocation in several ways. Evaluating each conception's ability to fulfill the allocative role could help us choose between them. While some believe subsidiarity should identify candidate authorities and the issues/subjects over which they should possess authority, 43 subsidiarity would be valuable if it could only allocate authority over particular issues/subjects amongst groups previously identified as having good claims to some authority. Recall the case of a province and a city with equal self-determination-based claims. If subsidiarity-based allocations are consistently plausible, subsidiarity beneficially resolves a problem common to all countries. If it cannot plausibly allocate authority, we may seek an account that can fulfill another moral role.
Subsidiarity could guide authority allocation at several stages in a country's development and in several ways, but all uses of subsidiarity for authority allocation purposes seek to establish a presumption favouring decentralization. Subsidiarity could, for instance, guide decisions about how to allocate powers

92
Da Silva at constitutional moments (e.g., an initial constituent assembly, moments of reform) or when entities decide when to devolve their powers to other entities. Realpolitik proponents may question whether abstract principles actually influence either type of decision. Yet decision-makers in both contexts normatively should-and often are politically required to-provide reasons for their decisions. Subsidiarity is at least proffered as a reason for decisions in both relevant contexts, and subsidiarity's advocates claim that subsidiarity can create a default presumption for local control over many issues/subjects. 44 Even if, moreover, decision-makers did not appeal to principles to justify their decisions, a set of normative principles we can use to judge allocative decisions remains desirable. Subsidiarity is at least a good candidate for that role. It may also play a role in evaluating how entities understand their powers-for example, highlighting how the powers should be understood as necessarily enmeshed with decision-making at other levels-or create a presumption of common devolution of those powers to more local levels. 45 Subsidiarity could also, and purportedly does, guide judicial decision-making about constitutionally-entrenched powers. 46 Courts use the principle to resolve jurisdictional disputes to clarify explicit allocations of authority and to fill jurisdictional gaps. One can examine whether subsidiarity provides adequate guidance on how to do so and whether invoking it actually provides intuitively compelling allocations to local authorities. Even if one is primarily interested in how to allocate authority for constitutional divisions of powers or devolution, literature on judicial decision-making helps clarify how the concept can be understood in practice and offers another way in which subsidiarity could guide decisions. I accordingly discuss some judicial decisions even when focused on other allocative moments below.
I make no assumptions here on whether an inability to allocate authority within countries will challenge its ability to fulfill other tasks. The question 'What makes entities candidate authorities?' is largely taken as answered to explore subsidiarity's allocative potential. The next section (surprisingly) demonstrates that no form of subsidiarity as such provides proper allocative guidance but attending to its forms/justifications clarifies its potential and limits. It demonstrates that arguments for subsidiarity either fail to provide specific, definite guidance on how to allocate powers; provide guidance that is unintuitive and at odds with prima facie acceptable existing authority allocations; or only present plausible guidance at the expense of subsidiarity's most fundamental commitments or collapsing it into other principles. To be clear, my concern is not that subsidiarity is not a panacea. Rather, my concern is that no plausible specification of subsidiarity fulfills (at least one of) the concept's primary task(s): providing normatively acceptable guidance on how to allocate authority based on a defeasible presumption of local control. This is so of each form of authority allocation in the preceding paragraphs.
Further detailing different conceptions of subsidiarity that aim to guide authority allocation within countries and claim consistency with liberal-democratic norms clarifies the issue. For instance, Andreas Føllesdal's prominent EUinfluenced view holds that "powers or tasks should rest with the lower-level sub-units : : : unless allocating them to a higher-level central unit would ensure higher comparative efficiency or effectiveness in achieving" state aims. 47 Others remove or add features. Hirschl does not appeal to comparative efficiency or effectiveness; his version "refers to the norm favoring the preservation or allocation of power to local authorities." 48 N.W. Barber requires allocation to those 'most affected' by a policy, not the 'lower' unit alone: subsidiarity "does not just embody a preference for smaller units over large ones: it allocates powers to the states containing the people who will be affected." 49 Still others connect EUinfluenced and earlier 'communitarian' conceptions focused on protecting all 'local' groups, including trade unions and religious groups. John Finnis's subsidiarity principle holds "that it is unjust for a higher authority to usurp the self-governing authority that lower authorities, acting in the service of their own members[,] : : : rightly have over those members." 50 These definitions share features (permitting provisional definitions) but disagree on core issues. Føllesdal echoes Barber in invoking a most affected principle. 51 Yet authors disagree on subsidiarity's scope of application. While the EU views it as applying in areas of "shared" authority, 52 it is also offered as a principle for allocating unique powers in the first place. 53 While Barber characterizes subsidiarity as a principle of 'constitutionalism' in the sense that it applies to all countries at all times and should be implemented in the institutional structure of all 'constitutional' countries, 54 others treat it as an allocative option. 55 47. Føllesdal, supra note 4 at 190. See also Føllesdal, "Federalism", supra note 5; Carozza, "IHRL", supra note 3; Allard-Tremblay, supra note 31. 48. Hirschl, supra note 3 at 221. 49. Barber

94
Da Silva Subsidiarity's relationship to federalism and self-determination also differs across accounts, leading to different outcomes. 56 Subsidiarity is also alternatively discussed as creating a presumption in favour of 'local' control, 57 a burden that must be met before powers can be allocated to more central authorities, 58 a duty to "aid" others in exercising their powers, 59 and/or a duty to assist them in fulfilling their self-defined ends. 60 Views on what can defeat subsidiarity's presumptive decentralizing tendency vary. Even Føllesdal notes that his "comparative efficiency" condition can be understood as a necessity condition or an effectiveness condition. 61 Whether federal governments are justified in acting when provinces will not act or only when they cannot do so also arises in the constitutional law of countries adopting subsidiarity or functionally equivalent principles. 62 Moreover, while extant definitions view subsidiarity as a decentralizing principle favouring allocating powers to the most 'local' unit possible, they may not always establish subsidiarity as decentralizing. The Treaty on European Union states that decisions should be made "as closely as possible to the citizen." 63 This could make decentralization a necessary component of accounts of subsidiarity that seek to reflect EU law. 64 Yet subsidiarity may not decentralize if the burden for overcoming its presumption is low. 65 Many powers could be allocated to federal governments due to comparative efficiency. Indeed, subsidiarity centralizes power in existing countries' legal practice (at least when applied to 'residuary' powers). 66 Subsidiarity's purported status as a decentralizing principle likely stems from frequent 'justifications' based on purportedly decentralizing (sub-)principles. Arguments appealing to those most affected by an issue/subject, democratic concerns, and/or unique local interests or knowledge each purport to justify decentralization. 67 The 'most affected' principle is challenged for being too vague 56. Cf ibid; Barber, supra note 5 at 215; Canadian caselaw in sources cited at supra note 5. 57. See Allard-Tremblay, supra note 31 (also stressing the burden needed to overcome it); Trevor Latimer, "Against Subsidiarity" (2018)  (viz., it is often unclear who is most affected by what issues, let alone subjects), 68 but its emphasis on stakeholder status appears important. 69 Stakeholder status also grounds democratic arguments for decentralization and subsidiarity: 70 local self-governance purportedly ensures the greatest representation possible in decisions affecting one (e.g., under subsidiarity, decisions are made by the group in which one's vote has its greatest impact). 71 Protecting unique local interests is likewise proffered to justify decentralization and subsidiarity. Subsidiarity is sometimes discussed as an efficient way to sort interests. 72 It may best protect unique interests. 73 The EU version is purportedly justified by a need to protect local interests and the defeasible presumption that 'local' groups best protect them. 74 Arguments for decentralization speaking to the unique interests and challenges of local entities or epistemic value of local knowledge could also support subsidiarity if the 'closest' groups can be expected to protect the interests or are epistemically well-positioned to address given issues/subjects. 75 Justifications for non-EU-based conceptions of subsidiarity likewise appeal to decentralization. Two 'liberty'-based views are representative. 76 The 'confederalist' view highlights the need for sub-state groups to whom citizens remain loyal as 'bulwarks' against tyranny. 77 Loyalty to a sub-state group of a sufficient size counter-balances central power, justifying local control to protect against unchecked central authority. 78 The existence of too many groups will leave no group strong enough to check power, requiring allocation only to some groups, but a subsidiarity principle limited to a small number could protect against tyranny. 79 The 'Calvinist/communitarian' view alternatively posits that local groups have unique values that must be protected from state interference. 80

Da Silva
suggests that these communitarian goods are so important as to justify subsidiarity even if it generates inefficiencies. 81 Subsidiarity, then, admits several forms grounded in compelling (sub-)principles. Nearly every form produces many prima facie plausible results, explaining why, for example, military and foreign policy powers often belong to central/federal governments while zoning powers often belong or are delegated to 'local' units. 82 If one form best allocates powers, it best fulfills subsidiarity's primary task. This will resolve the authority allocation problem and identify subsidiarity's best conception. Unfortunately, as we will now see, each modern conception of subsidiarity leaves important allocative questions unaddressed or relies on other principles to justify their conclusions, undermining the case for subsidiarity itself as an allocative principle.

The Case Against Subsidiarity as a Genuine Allocative Principle
Subsidiarity is prima facie compelling and has broad support, but its appeal rests on underlying (sub-)principles such that "apparent consensus on it has been gained only by obfuscation." 83 Subsidiarity shorn of its underlying justifications is a mere stipulation. 84 Using it as an allocative principle requires attending to its justifications and accepting its institutional implications. Yet existing conceptions underdetermine authority allocations. Even plausible allocative decisions that subsidiarity could generate rest on underlying principles that do the real allocative work. They may not make local control a presumptive norm and could even lead to centralization. Subsidiarity does produce centralization in practice. Subsidiarity, then, is not merely an incomplete allocative principle. 85 It is not a standalone allocative principle at all.
The first issue with subsidiarity as a standalone authority allocation principle relates to familiar boundary-based considerations. Subsidiarity does not provide univocal, consistently compelling understanding of 'closest'/'most local'/'most 81. Per Føllesdal, other historical conceptions, like the Catholic one, likewise speak to local interest protection. See Føllesdal, supra note 4. This helps explain why Cahill, supra note 4, denies the existence of a new 'modern' conception. Yet the conception found in modern constitutional, EU, and international authority at least purports to break from tradition and address a different question. The Catholic conception often addresses questions related to sphere sovereignty, which may not be identical to the authority issues addressed here-though, as discussed above and below, it may offer a better understanding of a unique role subsidiarity can play in moral theory. affected.' 'Close' and 'local' can be geographically-defined. Yet the plausibility of municipal control suggests that subsidiarity as a geographically-defined principle should presumptively favour cities. 86 Even if the principle need not extend further to neighbourhoods, which some contest, 87 this account struggles to explain the lack of municipal constitutional powers in most federations or the inconsistent history of delegation of authority to cities, which suggests that even historical decisions to provide (revocable) powers to cities are not viewed as even presumptively compelling. 88 An often-overturned 'presumption' favouring the 'most' local is barely a presumption. While one may argue that this is just evidence that existing allocations do not meet basic theoretical standards, the large explanatory gap remains problematic absent some heretofore-absent explanation of why present realities seem intuitively problematic grounded in subsidiary-related concerns.
'Closeness' is, moreover, difficult to specify absent a potentially undermotivated geographical stipulation. 89 This is a clear issue for accounts of subsidiarity that require group identities to track boundaries. 90 It is likely an issue for any account where allocative principles should be able to address real claims with some plausibility. Claims by non-territorial groups on self-determination-based grounds cannot be rendered implausible by geographical fiat. Even geographical boundaries can be split further. Maria Cahill's "primary" groups possessing presumptive decisional authority include non-governmental associations, like boxing clubs. 91 While clubs need not have authority if there are other reasons to bar them from possessing it (e.g., illiberalism), concerns about how to specify the 'local' raise questions about whether a presumption should favour greater authority for all 'local' groups. Subsidiarity at least struggles to explain existing authority allocations on any account that identifies anything 'closer' than provinces as candidate authorities. Border-drawing concerns will often make it difficult to apply.
Appeals to the 'most affected' entity require greater specification of the domain of an issue or subject and may not present a uniformly compelling answer favouring any entity, let alone presumptively favour decentralization. This concern is most acute with respect to division of powers-based concerns regarding authority over subjects. It is hard to establish the entity most affected by immigration or healthcare. 92 COVID-19 demonstrated public health policies' differential impacts, but the entity 'most affected' by public health law is non-obvious. 93 A most affected principle could select two entities. It may not even provide clean 86. See municipal authority-focused sources, including those in supra notes 13, 14. 87. See e.g. Jerry Frug, "Decentering Decentralization" (1993) 60:2 U Chicago L Rev 253. 88. Weinstock, "Cities", supra note 12 is inspired by many states' failure to provide powers to cities.

98
Da Silva answers in otherwise-easy cases favouring local control. 94 Consider municipal highways/roads. Should commuters have more say on a related policy because they use roads (supporting provincial control) or less because they do not pay property taxes? 95 'Most affected' principles underdetermine this point. This problem extends to authority over issues. Consider a provincially-owned/-operated park within city limits but most often used by suburban residents. Who is 'most affected' by decisions to close that park to protect public health? Subsidiarity alone provides no answer. These problems do not arise only because cities or sub-state nations have prima facie compelling claims. The claims' prima facie plausibility establishes the need for a further allocative principle and difficulties addressing the claims are problematic, but the larger issue is that subsidiarity alone has difficulty explaining when it should apply. Perhaps cities, sub-state nations, et al. will never justifiably receive power, or one can justify only providing authority to central, federal, or provincial governments. Subsidiarity could be useful where the only entities recognized as even candidate authorities are federal and provincial governments (though this would tie subsidiarity's plausibility to a substantive response to the ontological question and limit its application to a small number of cases, retaining the need for another allocative principle). Even then, difficulties identifying which powers must belong to more 'local' groups and when and how the presumption can be overcome would remain. Whether subsidiarity as such can plausibly set the relevant standards is at best debatable (as further demonstrated below).
Even stable meanings of 'closest'/'most local'/'most affected' do not uniformly favour subsidiary entities as authorities over discrete decisions, let alone constitutional powers. It is non-obvious that, for example, 'local' cities should presumptively possess many powers. Arguments for municipal control over immigration and healthcare powers remain controversial. It is difficult to accept that the burden is on others to disprove them. Similar problems occur even where we only analyze federal and provincial governments. Whether national or local cultures are most affected by immigration decisions is debatable. 96 Mere stipulations that, for example, provinces should control healthcare policy are difficult to justify, especially given distributional inequities across provinces who possess it. 97 While provincial control may be best in both cases, a stipulated presumption cannot justify provincial control. Treating subsidiarity as a principle for allocating authority over issues, rather than subjects, cannot avoid these concerns. get tested for an illness during a pandemic may not make cities even epistemically-best-positioned to place them. Even if one treats subsidiarity's purported justificatory sub-principles as fundamentally decentralizing, they may not uniformly allocate powers to the most local entity. Sub-principles include a most affected principle, democratic principles, principles stressing the import of unique local interests, challenges, or knowledge, and/or principles appealing to the need for bulwarks against tyranny. Many cannot identify which 'local' entity should possess which powers. Parallel arguments for provinces and cities should lead to more municipal powers on a subsidiarity-based view favouring most local control. 98 Yet subsidiarity's justificatory sub-principles may not produce this result. Even appeals to combinations of sub-principles could produce justificatory ties. For instance, both entities may secure equal 'loyalty' from members and be equally knowledgeable about the implications of a decision related to whether the parks above should be closed. Considerations that could break ties, like institutional capacity, are not always available. Available considerations likely favour provincial control, defeating presumptions favouring control by the most 'local' city, and again suggesting that subsidiarity is at best a strong candidate authority for allocating powers between federal and provincial governments alone. Even then, the decisions are likely only plausible regarding issues, not subjects. Strong cases for 'local' control over subjects, like land management or mining, will (at best) prove rare.
Issues like those undermining 'most affected' principles above also apply to 'democratic' and 'local interest' arguments. 99 Necessarily brief comments on each motivate general problems.
Some democratic concerns stem from boundary problems. It is difficult to determine the scope of subjects or most affected parties, making it hard to identify who should have what level of representation in decision-making procedures. Other concerns stem from democratic approaches' lack of clear guidance on which powers should belong to which entities and/or how they provide intuitively problematic guidance. Proponents of municipal authority highlight how cities face unique challenges (re: diversity, homelessness, criminality, etc.) and are uniquely affected by existing federal policies (re: migration, etc.); their lack of decision-making authority in responses to those challenges and/or development of policies is worrisome. 100 Yet cities and sub-state nations are uniquely affected by many considerations. Cities face unique issues due to traditional features, like population density, and to recent developments, like greater immigration into cities (and, then, development of local cultural groups) and demographic 'sorting' whereby cities tend to be more 'liberal' than rural communities. 101 Their lack of authority over related matters arguably raises democratic issues, which can be acute where cities are more liberal and institutional considerations lead to more 98. See supra notes 14, 15. 99. See Latimer, supra note 57, suggesting that major arguments also face empirical challenges. 100. See Hirschl, supra note 3 at 224, building on Weinstock, "Cities", supra note 12. 101. See Hirschl, supra note 3; Weinstock, "Cities", supra note 12; de-Shalit, supra note 14.

100
Da Silva conservative federal and provincial governance. 102 If demographic sorting leads cities to consistently 'lose' on matters that uniquely impact them, cities likely lack adequate representation on the matters that most impact them. 103 Parallel arguments apply to sub-state national groups. 104 Yet democratic appeals to subsidiarity struggle to explain which powers must belong to cities or sub-state nations and likely cannot provide intuitively compelling results. Whether 'local' groups' experiences should require municipal control over immigration policy is at best debatable. 105 If one accepts that result, problem cases remain. For instance, higher crime rates in cities suggest that criminal law 'most affects' city-dwellers. 106 A presumption favouring primary municipal authority over crime is uncompelling. The same issues arise with appeals to local interests, challenges, or values divorced from democratic concerns. These principles provide some guidance on when cities or sub-state nations should possess particular powers. If, for example, one is best-positioned to protect a local culture, providing it with powers over cultural policy is intuitively compelling. Yet the scope of necessary cultural protections is difficult to parse. A presumption that it should, for example, control all education policy is hard to justify on local interests alone given education's broad societal impact. Moreover, the range of unique interests, challenges, and values is wide enough to again generate justificatory ties between candidates. Entities could, for example, have equal self-determination-based claims for controlling language policy. A self-determination-based approach to subsidiarity would then be at best an incomplete allocative principle. Finally, even if one could identify which entities have unique interests, challenges, or values, it is hard to know whether and when control over which subjects is necessary to complete them. This could change over time. 107 These 'justificatory' (sub-)principles too at best allocate authority over issues.
Denying that powers should be provided to cities or sub-state nations cannot avoid these concerns. For instance, Loren King discusses parallel arguments for provincial and municipal authority but argues that subsidiarity should not apply to cities. 108 Yet an issue remains even if we grant the substantive ontological account generating that result: King jettisons the presumption favouring most local control and limits subsidiarity's application more narrowly than one would desire of a complete allocative principle. Subsidiarity alone cannot explain why cities and sub-state nations' prima facie compelling claims should not be realized if it is justified using features that establish those claims. Another allocative principle remains necessary. 109 102. See Rodden, supra note 10. 103. Ibid. One may argue that that municipal bodies should set the terms of their powers, but claim that is too strong. Moreover, as discussed later in this section, subsidiarity does not protect municipal control over democratic process in practice. Epistemic arguments better guide when 'local' entities should make decisions but still provide incomplete guidance in division of powers cases. What qualifies as knowledge is contentious and whether more 'local' entities consistently have better knowledge in a way that generates a presumption favouring local decision-making is questionable-local governments may often lack knowledge of a decision's position in a larger decision-making context and impact elsewhere-but there is, at least, a decision procedure for identifying when local entities should make decisions on epistemic approaches. I cannot, however, accept that local entities generally have more knowledge about policy areas. I do not even know how to judge such claims about epistemic authority over subjects. Even epistemic approaches thus have limited scope. Moreover, epistemic positions may not sustain over time. Even if cities are best-positioned to decide where to place testing units for a particular virus on a particular date, they are not obviously best-positioned to handle public health policy generally and their knowledge may not sustain and warrant long-term delegation over narrow health facilities placement issues. 110 'Bulwarks'-based arguments explain why most powers should belong to provinces but may not do so in a uniform manner and limit subsidiarity's scope of application more than proponents desire. They hold that non-central/federal powers must be allocated to groups that sustain loyalty over time and do not eliminate concurrent loyalty to the country or splinter authority among so many groups so as to leave none with enough power to be bulwarks. 111 While some think these conditions only apply to provinces, 112 empirical evidence suggests people maintain loyalty to cities and sub-state nations over time and can do so while maintaining commitments to a 'state' identity. 113 Extending authority to many entities may limit any entity's ability to do so, but three or four levels of governance, including municipal or 'sub-state national' levels, could act as bulwarks. 114 Questions about how to allocate powers between provinces and cities, for example, thus remain. Resolutions may not favour 'most local' control. Moreover, bulwarks-based arguments alone do not address which powers 'lower' authorities require to be bulwarks. 115 Even if we could resolve those problems, in turn, the idea that our allocative principle would only apply to central, federal, and provincial governments would still leave it with an undesirably narrow scope, retaining the need for another principle. 116 110. I build on Da Silva, supra note 11 here. 111. See e.g. Somin, supra note 10; Levy, supra note 53. 112. See Levy, supra note 53; King, supra note 4. 113. See supra notes 14, 15. 114. On the possibility of three or four levels of governance, see e.g. Barber, supra note 5 at 194 n 24. 115. Ibid at 191 notes this problem occurs even before bringing in cities. Moreover, other arrangements may be better bulwarks; see Latimer, supra note 57 at 289-90. 116. Sophisticated variants also face issues. For example, Allard-Tremblay's epistemic account provides powers to entities best-positioned to address 'common' goals. Conditions requiring 'common actions' specify when a presumption that the most 'local' entity can be defeated. Yet such goals are often unidentifiable. Allard-Tremblay's value pluralism bars his stipulating them or what groups are most 'local.' Similar issues explain why he views subsidiarity's benefits are largely discursive and does not focus on direct allocative guidance. See Allard-Tremblay, supra note 31.

Da Silva
Examining each possible justification, then, challenges any case for subsidiarity as a genuine allocative principle. Any presumption favouring local control is likely only plausible regarding issues, not subjects. This is especially bad for the efficiency-based view, which is supposed on contingent realities and needs constant updating. 117 But the problem generalizes: No subsidiarity principle appears to provide uniformity on decisions over subjects, leaving questions about, for example, immigration and healthcare reform, or what to do in constitutional moments where subsidiarity is supposed to be most compelling, unresolved. This claim does not make the "mistake" of viewing subsidiarity as mere efficiency. 118 It just recognizes subsidiarity's limits.
These considerations would be less forceful if subsidiarity consistently guided judicial decision-making in a compelling manner, but problems recur when applying the principle. One may contend that nothing in the forgoing undermines judicial use of the principle for authority allocation purposes. 119 Judicial application of the principle could vindicate its utility in concrete cases. My claim is that subsidiarity fails to create a plausible theoretical presumption favouring local control, but the point underlying this critique is fair: empirical data on the use of the principle is important on any plausible view. However, judicial use of subsidiarity does not consistently support its real-world plausibility (if it does so at all). To wit, European law heavily relies on national parliaments to enforce the subsidiarity principle through political pressure and European judicial treatments of subsidiarity amalgamate points from multiple distinct intellectual traditions, each of which raises the analytic problems undergirding critiques above. 120 While domestic courts invoke the principle elsewhere, in turn, their results are hardly inspiring. Canadian jurisprudence is instructive. The Constitution of Canada does not contain a subsidiarity principle. 121 This alone raises a challenge for many countries' use of subsidiarity: it is often unclear whether and when judges can validly invoke it. 122 Yet scholars suggest that the federal government of Canada's power to legislate with respect to issues of 'national concern' should be understood as reflecting a deeper constitutional concern with subsidiarity and that subsidiarity itself is recognized as a valid interpretive principle in Canadian constitutional law. 123 Extant jurisprudence admittedly provides some support for both claims. However, the same case law suggests that subsidiarity is not an action-guiding principle for resolving jurisdictional disputes that can provide intuitively compelling results favouring presumptive local control.
With respect to issues of national concern, Canada's federal government possesses a power to legislate to ensure Canadian "Peace, Order, and good Government" (POGG); this is a 'residuary' power to legislate only where there is a gap in jurisdictions (e.g., neither federal nor provincial governments can explicitly legislate in the domain), in cases of national emergencies, and over issues of national concern. 124 Subsidiarity proponents suggest that this framework operates in a subsidiarity-like manner since the federal power only exists where provincial authority proves inadequate to the task at hand. 125 Yet even if we accept that contentious reading, we will now see that it does not appear to support practical uses of the subsidiarity principle. 126 Canadian judges have long struggled to develop a clear test for identifying matters of national concern and recent legal developments do not support the use of subsidiarity-like principles. 127 Historical tests were "difficult to apply." 128 The most recent re-articulation of the test, in the References re Greenhouse Gas Pollution Pricing Act may not clarify matters, maintaining the basic problem. 129 Under that test, the national concern branch of POGG can be validly invoked where there is (i) a "matter is of sufficient concern to the country as a whole to warrant consideration as a possible matter of national concern" that has (ii) "singleness, distinctiveness, and indivisibility" (SDI) that distinguishes it provincial 'inability,' let alone which of these should suffice; the reference includes material supporting multiple interpretations of this criterion. The majority judgment also does not resolve historical questions about the meaning of 'reconcilability.' 138 It only offers two paragraphs on how to understand that final stage. 139 There are, then, reasons to question whether Canadian national concern doctrine is properly action-guiding. The result in References re Greenhouse Gas further highlights how the branch itself does not reflect a meaningful presumption of local control. That case concerned a federal emission standard. All parties agreed that the provinces could set their own standards. Several chose not to do so. There was accordingly no sense in which the provinces were literally unable to set the relevant standards. However, the Supreme Court of Canada permitted strong incursions into provincial jurisdiction in the name of national concerns. While an Alberta Court of Appeal decision appealed to the subsidiarity principle to highlight the constitutional problems with the act in question, 140 the Supreme Court of Canada did not even directly engage with subsidiarity arguments. 141 It instead permitted strong incursions into admitted areas of provincial control, which combined with federal paramountcy rules to aid centralization. 142 Any purported presumption favouring local control is, then, apparently very weak. I doubt its existence.
Subsidiarity as a standalone interpretive principle does not fare much better in Canadian constitutional law. 143 Many references to the principle that scholars use to highlight its importance were mere obiter and did not explain how the principle should apply. 144 Even the most famous case referencing the principle, which upheld a municipal anti-pesticide law while highlighting the need for tri-level governance, was decided on the grounds that the municipal law fell under the terms of a valid provincial law, rather than the importance of local control as such. 145 The first detailed analysis of the subsidiarity principle in Canada was, in turn, part of a 4-4 split with minimal force; it again relies on provincial authority for minimal municipal powers and explicitly states that subsidiarity cannot alter the division of constitutional powers. 146 More recent attempts to invoke subsidiarity to save local laws largely falter. 147 Appealing to subsidiarity did not, for example, save local environmental standards that would impede a federal undertaking (viz., a pipeline), undermining the suggestion that a concern with environmental regulation best explains References re Greenhouse Gas: centralization seemingly trumps environmental concerns in practice. 148 While scholars also argue that democracy-and local interest-based subsidiarity arguments should require that provinces defer to municipalities on the drawing of electoral boundaries, the Supreme Court of Canada recently held that provinces can unilaterally change districts mid-election. 149 Subsidiarity can bolster some attempts to save bylaws in rare cases, 150 but the case law does not consistently favour local control. It also may not track intuitions on appropriate local control given its limits on provincial and municipal powers.
If one takes a more sanguine view on the theoretical and jurisdictional issues above, subsidiarity is still at best able to address delegation-based and divisions of powers concerns on a question-by-question basis, leaving large holes in subsidiarity's ability to serve its intended allocative function. But even a uniform 'presumption' in discrete decision-making contexts is unlikely to hold. It is not clear that we should presumptively defer to cities on questions about where to place testing centres within municipal boundaries or when to close a park. Principles that could justify municipal control, including democratic and local control-based arguments, may equally favour provinces. Whether any tiebreaker principles could uniformly establish a presumption favouring subsidiarity as a basic allocative norm is at best unknown. Appeals to those 'most affected,' democratic concerns, or local interests or knowledge are unlikely to frequently allocate to the 'closest' level of governance if it is defined territorially-for example, most stakeholders will seek to move many powers away from cities-and such appeals to those 'most affected' or territoriality likely beg important questions anyway. Attempts to limit the scope of the analysis to central, federal, and provincial governments may not avoid this result and would limit subsidiarity's scope of application in a way that still requires another allocative principle.
Where, in turn, 'justifications' for subsidiarity favour a 'local control' presumption, other (sub-)principles do the allocative work, making subsidiarity (at best) a mechanism for weighing other concerns. If, for example, communitarian ties justify presuming 'local' control over particular issues and identifies areas-for example, value-laden areas that are unnecessary for wider state stability, possibly including cultural policy, healthcare, or (more controversially) immigration powers-as best belonging to cities or sub-state nations, the community values principle guides allocation, not subsidiarity. Likewise, appeals to subsidiarity do no extra work where analytically severable democratic principles plausibly justify general allocative decisions. 151 If we are interested in democratic considerations, protecting local interests, leveraging special local knowledge, or bulwarks, we should appeal to those considerations to justify claims and weigh their value against other interests. Appealing to subsidiarity adds nothing to this weighing process.
The issue, then, is not merely that subsidiarity relies on other principles for its purported value. Rather, appeals to 'subsidiarity' either lead to implausible conclusions by the light of those very underlying principles or add nothing distinct to the moral decision-making process. Appeals to subsidiarity may even obscure relevant moral considerations in practice. The Supreme Court of Canada did not need to engage with subsidiarity directly in a municipal bylaw case since subsidiarity is only an interpretative principle. 152 Yet democracy is an even more fundamental principle of Canadian constitutional law. 153 One wonders if concerns about subsidiarity distracted from the important democratic concerns that may have been more important in the case. We cannot know for sure but even that possibility should be worrying.
Ultimately, then, subsidiarity is not a unique allocative principle and cannot serve its primary intended function. This does not mean that subsidiarity lacks any value but makes it less valuable and stresses the need to reflect on other allocative principles, such as the self-determination or epistemic value principles, to see if they play distinct allocative roles as claimed. Even if subsidiarity does not add meaningful content to discourse, it may, for example, have an important rhetorical role. For instance, some believe that subsidiarity's inability to answer all allocative questions is unproblematic because invoking it occasions reflection on important moral concerns (e.g., the justificatory sub-principles above) and leads to more informed allocative decisions. 154 Appealing to subsidiarity might contribute to legal discourse on authority allocation, which could itself serve valuable ends, like building community in a polis and strengthening the legitimacy of decisions therein. 155 Simply adopting subsidiarity could also promote stability in the face of deep disagreement within countries. 156 Yet these cases for subsidiarity are practical (viz., non-theoretical), empirically contestable, and contingent. Moreover, 'legal discourse' must be ongoing to accrue the benefits. Ongoing constitutional debate is rare, again suggesting that subsidiarity will at best allocate authority over issues. Another allocative principle thus remains necessary on this best-case 'discursive' view. If one finds subsidiarity compelling, I recommend appealing to the underlying principles that make it seem compelling in a case instead. They provide more direct, normatively distinct guidance.
One still should not reject an otherwise valid claim in real-world cases simply because it is couched in subsidiarity language, but we should lament it if valid claims must be couched in that language to succeed. For instance, appeals to subsidiarity have strengthened Indigenous authority claims in Canada in the face of federal matrimonial property laws that would have otherwise applied, thereby serving a community-building role and helping local communities better protect individual rights than the federal government. 157 That case suggests that subsidiarity can be rhetorically valuable for fostering Indigenous self-governance and protecting universal rights in local settings. One of these underlying concerns may warrant appealing to subsidiarity despite its deficiencies as a standalone authority allocation principle capable of fulfilling its intended ends. I do not think we should remove a valuable tool from real-world battles for justice. Yet selfdetermination, anti-colonialism, and individual human rights protections could justify the Indigenous matrimonial property law absent an appeal to subsidiarity. It is a shame that the relevant communities needed to appeal to a problematic subsidiarity principle if their case for authority was already over-determined. I would prefer that direct appeals to the principles underlying subsidiarity's seeming plausibility had more rhetorical force. This would permit more direct weighing of relevant moral interests. But public discourse need not mirror philosophy seminars.

Conclusion
Subsidiarity cannot serve one of its primary intended roles: allocating authority within countries. Its allocative potential is at best limited to cases where there is shared or no clear jurisdiction over issues, and relies on justificatory sub-principles. A 'presumption' favouring local control based on those principles likely cannot withstand scrutiny, and principles other than subsidiarity to do the allocative work when allocating authority to more 'local' entities is justified.
Testing subsidiarity's potential as an allocative principle nonetheless furthers discussions in several literatures. Regarding subsidiarity, while the basic concept could be valuable even if it cannot identify all the entities that have prima facie compelling claims to some authority, the forgoing suggests that 'subsidiarity' must serve a function beyond guiding authority allocation in countries to be valuable. One may still argue that this merely creates a burden for future theories of subsidiarity to provide better allocative guidance. The way some conceptions of subsidiarity and underlying principles seemed more compelling than others suggests some paths are more likely to be fruitful and provides reason to prefer those conceptions. Yet even the best conceptions appear unable to serve as wide-scope allocative principles. For instance, while epistemic considerations may do a good job of allocating powers, questions remain about how to specify greater knowledge, and epistemic accounts appear to justify wider municipal powers than many are willing to accept. Limiting the principle to federal and provincial governments may better explain observed prima facie justified allocations but limits subsidiarity's application more narrowly than proponents desire. Even then, epistemic value, not subsidiarity, is the guidepost. Absent more compelling accounts of subsidiarity's allocative value than those above, then, theorists should (surprisingly) no longer view an allocative function within countries as (one of) subsidiarity's key moral role(s). If subsidiarity is to be more than a structuring framework for democratic decision-making, conceptions that fulfill other moral functions are preferable. Subsidiarity may, for instance, still be valuable as a means of allocating authority at the international level or explaining why associations, like guilds or boxing clubs, are ontologically valuable. 158 Conceptions capable of fulfilling these functions will be preferable.
Regarding authority allocation, the forgoing highlighted the need for a new allocative principle and desiderata thereof. The principles undergirding subsidiarity not only engender discussion, 159 but demonstrate that deferring to central/ federal governments is also likely unwarranted as a general rule. Considerations explaining subsidiarity's possible allocative value may not justify a presumption favouring control at the most 'local' level possible, but they are important moral concerns and likely equally defeat presumptions that uniformly favour central/ federal governments. Practically, this finding provides reason to question central or federal governments who rarely delegate any authority and divisions of powers that recognize few provincial powers. It also undermines paramountcy, a principle under which federal laws always supersede provincial ones within predetermined areas of shared jurisdiction. 160 It is simply not obvious that federal decisions will always be 'better' in any or all relevant respects.
Finally, regarding method, the need for allocative principles to address real political claims-and ways subsidiarity (and now paramountcy) cannot uniformly and plausibly allocate authority within countries-stresses the need to examine whether other principles can supplement or replace subsidiarity and more plausibly allocate authority. An acceptable principle should guide each allocative decision above in a way that resolves the cases motivating this work, including conflict cases, and explain what justifies providing specific powers to specific entities (and when it does so). Perhaps no one principle can set general presumptions about how to justifiably, let alone ideally, allocate authority within countries. Decisions may need to be made through democratic processes. 161 Even then, we should seek a small set of principles for evaluating democratic

110
Da Silva decision-making. Allocations that appeal to features that made entities ontologically capable of making plausible authority claims benefit from simplicity. But separating ontological and allocative concerns permits necessary variation in the decisions.
The task moving forward is to identify whether any candidate allocative principles lack subsidiarity's issues. If every principle faces issues like those above, the case against subsidiarity will admittedly have less purchase. One should not, however, assume that result ex ante. The preceding provided tools for assessing candidate principles and suggested that some which purportedly support subsidiarity are more plausible candidates than subsidiarity itself. The next step is to determine whether they and other principles can survive the kind of scrutiny above.