The Fiduciary Relationship

Paul B. Miller

Forthcoming in: Andrew S. Gold & Paul B. Miller, eds., Philosophical Foundations
of Fiduciary Law (Oxford: Oxford University Press)

Introduction

The conceptual structure of private law is, to a considerable extent, formal in nature.

Principles of private liability are mediated by relatively stable forms of action,

interaction, and organization. The formal character of private law is reflected in the fact

that the application of liability rules is often preconditioned on judicial determination that

the interaction of the plaintiff and defendant fell within a form (or kind) of action,

interaction, or organization to which certain mandatory or default rules apply. Thus, for

example, a simple claim for breach of contract turns on the existence of a primary

obligation of contractual performance, which is in turn rooted in a particular form of legal

relationship – i.e., a contractual relationship.1 The claim for breach of contract will fail

unless the claimant can establish that the underlying relationship was actually contractual

– i.e., that it possessed the formal properties (validity conditions) of a contract. Similarly,

a claim of trespass to real property turns on establishment of ownership or a lesser

interest in the real property (e.g., a leasehold interest) that generates a right of exclusive

possession. Here, too, we have a liability rule the application of which is conditioned on


Assistant Professor, McGill University Faculty of Law. I am grateful to participants in the 2013 Fiduciary
Law Workshop at Notre Dame Law School for their input. I am also appreciative of questions and
comments provided by participants at the Philosophical Foundations of Fiduciary Law Workshop held the
DePaul College of Law. For detailed comments, I am particularly grateful to Christopher Bruner, Evan
Fox-Decent, Andrew Gold and Julian Velasco.
1
The fact that there may be several variants on a given form, such as a contract, does not detract from the
formal character of liability. It is instead a reflection of the mutability and complexity of legal forms as
constructions of reason. On varieties of contract, see Hanoch Dagan and Michael Heller, Freedom of
Contracts, forthcoming.

Electroniccopy
Electronic copy available
available at:
at:https://ssrn.com/abstract=2353062
http://ssrn.com/abstract=2353062
the form of the relationship between litigants (owner and non-owner; leaseholder and

non-leaseholder). Modifications of, or deviations from, a rule may also be formally

determined (e.g., the nature of the relationship between landlord and leaseholder informs

rules governing the rightfulness of the former’s presence on leased property).

Many of the forms of action, interaction and organization in private law are well

understood or at least widely analyzed. However, that is not true of fiduciary law. There

is considerable uncertainty over the basis, nature and, scope of fiduciary duties as well as

their justification. These uncertainties necessarily raise questions about the structure of

fiduciary law.

Is fiduciary liability, like some other bases of equitable intervention, largely

discretionary?2 Or is it, at least in part, structured formally? Addressing this question

requires that one consider shared experience as reflected in doctrine and the practices of

lawyers and judges. To what extent, if at all, are we concerned with the general nature of

the relationship between litigants when determining whether fiduciary liability rules

apply? Supposing we are concerned with the general nature of their relationship, do we

treat it as distinctive? I shall presently argue that we are (as a matter practice), and should

be (as an analytical matter), very much concerned with the fiduciary relationship and its

distinctive, formal, properties.

Let us begin at a pre-critical level, with our habits of thought and speech. As

lawyers and legal scholars we are, I think it is fair to say, in the habit of referring to

people as fiduciaries, and offices and relationships as having a fiduciary character, with

2
As Laycock suggests in his essay on the influence of equity on the common law. See Douglas Laycock,
The Triumph of Equity, 56 LAW & CONTEMP. PROBS. 53 (1993). See also Peter Birks, Equity in the Modern
Law: An Exercise in Taxonomy, 26 U.W. AUSTL. L. REV. 1. (1996)

Electroniccopy
Electronic copy available
available at:
at:https://ssrn.com/abstract=2353062
http://ssrn.com/abstract=2353062
the implication that they are instances of a general kind or category.3 So, for example, we

say with confidence that lawyers, directors, agents, and trustees are fiduciaries, and

perhaps with some diffidence that parents, doctors, and public officials are (or should be)

so considered. Similarly, we say of offices that they, too, have a fiduciary character (and,

by extension, we treat occupation of an office as a reason for considering a person or

relationship to be fiduciary). Finally, we speak of persons and offices as being implicated

in fiduciary relationships. Hence, we say that lawyers have a fiduciary relationship with

their clients, that corporate directors stand in a fiduciary relationship to the corporation

they serve, and so on. On a critical level, the implication of kind-status is reflected

positively in efforts to identify common characteristics of fiduciaries.4 It is reflected

negatively in our opinions, beliefs and decisions (authoritative or otherwise) that certain

people are not and could not be fiduciaries.

These practices may simply reflect a shared intuition that there are salient

resemblances between particulars; resemblances for which a collection of loosely related

terms (fiduciary, fiduciary office, fiduciary relationship) provide suitable shorthand.5 But

given that these are technical legal terms, and our practices are the practices of lawyers, it

seems more likely that they reflect conceptual and practical exigencies generated by

customary usage. And in law, the usage implies more than ease of reference. To refer to a

person, office, or relationship as fiduciary is to suggest certain significant legal

3
Hereinafter, I shall speak of “kinds” and “forms” interchangeably as shorthand for general kinds or
categories of roles, offices and relationships
4
See generally TAMAR FRANKEL, FIDUCIARY LAW 4-6 (2001); and Laura Hoyano, The Flight to the
Fiduciary Haven, in PRIVACY AND LOYALTY 169 (Peter Birks ed., Clarendon Press 1997).
5
As suggested by John Glover, The Identification of Fiduciaries, in PRIVACY AND LOYALTY 269 (Peter
Birks ed., Clarendon Press 1997).

Electronic copy available at: https://ssrn.com/abstract=2353062

consequences. A fiduciary is subject to a range of distinctive and demanding legal

obligations.6

Given customary usage, the suggestion that the fiduciary relationship is an

indefinable construct may seem surprising. Nevertheless, this is conventional wisdom

amongst leading fiduciary law scholars. In Section 2 of this chapter, I introduce and

challenge this view by contrasting it with judicial practice and opinion. In Section 3, I

offer a theory of the fiduciary relationship – the fiduciary powers theory of the fiduciary

relationship (hereinafter: fiduciary powers theory) – centered on a definition of the

relationship. The viability of a definition is partly a function of its capacity to encompass

phenomena commonly understood as denoted by the defined term. Accordingly, in

Section 4 I show how the fiduciary powers theory explains the kind-status of a set of

relationships widely considered fiduciary. The utility of a definition lies in its capacity for

differentiation of phenomena and is best revealed where differentiation is hard due to

confounding similarities. Thus, in Section 5 I show how the fiduciary powers theory

enables principled resolution of disagreements over the fiduciary character of other

relationships. In Section 6, I offer some brief conclusions.

2. Conflicting Views on Fiduciary Relationships

Current thinking about fiduciary law features two prominent views on fiduciary

relationships. The dominant academic view is that the fiduciary relationship is

indefinable. The dominant judicial view is that the existence of such a relationship is a

precondition of fiduciary liability. These views conflict. Judges treat the fiduciary

relationship as conceptually central to fiduciary liability; leading academics deny the

6
See generally Peter Birks, The Content of the Fiduciary Obligation, 34 ISRAEL L. REV. 3 (2000).

Electronic copy available at: https://ssrn.com/abstract=2353062

coherence of the concept. This conflict is paradoxical. Judges condition liability on the

existence of a relationship that falls within a general kind or category that many

academics insist is illusory or incoherent. The paradox is troubling in itself but also

because it generates certain complexities for our understanding of fiduciary duties. Case

law predominantly suggests that fiduciary duties are responsive to something in the

nature of the fiduciary relationship, and thus that fiduciary liability is formally structured.

But leading academic opinion seems to imply otherwise.

A. The Fiduciary Relationship is Indefinable

The notion that the fiduciary relationship is indefinable is venerable.7 It has also proved

remarkably durable.8 Len Sealy was an early proponent of this view. In an article on

fiduciary relationships, Sealy averted to the “problem of defining a fiduciary situation.”9

He concluded that the problem is irresolvable: “The word “fiduciary” … is not definitive

of a single class of relationships to which a fixed set of rules and principles apply … the

7 I attribute this position primarily to academics because definitional reasoning is itself largely an academic
preoccupation. It is rare to find direct judicial pronouncements on point. But see Wilson J, dissenting in
Frame v. Smith [1987] 2 SCR 99 (SCC) 135, who noted that: “there is no definition of the concept
“fiduciary” apart from the contexts in which it has been held to arise.” Likewise, Brennan CJ in Breen v.
Williams (1996) 186 CLR 71 (HCA) 92 noted that “the law has not, as yet, been able to formulate any
precise or comprehensive definition of the circumstances in which a person is constituted a fiduciary in his
or her relations with another.” Fletcher Moulton L.J.’s famous warning in Coomber v. Coomber [1911] 1
Ch 723 about “unthinking resort to verbal formulae” in fiduciary law has been interpreted by some as a
condemnation of definitional reasoning. But I think that the best interpretation of the dictum in Coomber is
that given to it by Millett LJ in Bristol & West Building Society v. Mothew [1998] Ch 1. Millett LJ takes the
dictum to mean that greater care is required in definition of terms. Thus he began his analysis of a claim of
fiduciary breach involving negligence by stating that it is “necessary to begin by defining one’s terms.”
More specifically, it is necessary that we think about fiduciary law in terms of concepts that are distinctive
to it (though Millett LJ was concerned with identifying what is distinctive about fiduciary obligations, not
fiduciary relationships).
8
This is true, notwithstanding some significant attempts to define the fiduciary relationship or to clarify its
basic nature. See especially Ernest J. Weinrib, The Fiduciary Obligation, 25 U. TORONTO L.J. 1 (1975);
J.C. Shepherd, Towards a Unified Concept of Fiduciary Relationships, 97 L.Q.R. 51 (1981); and D.
Gordon Smith, The Critical Resource Theory of Fiduciary Duty, 55 VAND. L. REV. 1399 (2002).
9
L.S. Sealy, Fiduciary Relationships, CAMBRIDGE L.J. 69, 69 (1962).

Electronic copy available at: https://ssrn.com/abstract=2353062

mere statement that John is in a fiduciary relationship towards me means no more than

that in some respects his position is trustee-like.”10

An influential endorsement of this view was subsequently offered by Paul Finn in

his classic treatise, Fiduciary Obligations.11 Finn’s book was ventured on the basis of his

immediate and unequivocal rejection of the notion that the fiduciary relationship is a

distinctive and coherent kind of legal relationship. On the very first page of his book,

Finn proclaimed:

[I]t is meaningless to talk of fiduciary relationships as such. Once one looks to the
rules and principles which actually have been evolved, it quickly becomes
apparent that it is pointless to describe a person … as being a fiduciary … The
rules are everything. The description “fiduciary”, nothing.12

The views of Sealy and Finn have proven widely influential and are shared by

leading contemporary scholars. For instance, Deborah DeMott argues that fiduciary law

has “developed through a jurisprudence of analogy rather than principle” and insists that

“the law of fiduciary obligation is situation-specific.”13 The situation-specificity of the

law is such as to render definitional reasoning futile. DeMott argues that “the

characteristics of even the standard or conventional fiduciary relationships … are too

variable to enable one to distill a single essence or property that unifies all in any

analytically satisfactory way.”14

10
Id. at 73 (original emphasis).
11
P.D. FINN, FIDUCIARY OBLIGATIONS (1977).
12
Id. Interestingly, in his later writings Finn recanted, expressing worry over indiscriminate use of the word
“fiduciary”: “our present uncertainty is thought to be exacerbated by the lack of a workable and
unexceptionable definition of a fiduciary … Definition, however imperfect, has its place.” P.D. Finn, The
Fiduciary Principle in EQUITY, FIDUCIARIES AND TRUSTS 26 (T.G. Youdan ed., Carswell 1989).
13
Deborah A. DeMott, Beyond Metaphor: An Analysis of Fiduciary Obligation, 37 DUKE L.J. 879, 879
(1988).
14
Deborah A. DeMott, Breach of Fiduciary Duty: On Justifiable Expectations of Loyalty and their
Consequences 48 ARIZ. L. REV. 925, 934-935 (2006).

Electronic copy available at: https://ssrn.com/abstract=2353062

 Similarly, Matthew Conaglen argues that fiduciary duties can and ought to be

analysed independently of fiduciary relationships.15 He, too, recognizes that there are

categories of relationship in which “fiduciary duties are indubitably owed” but thinks

there is little to be gained in analysing them as a kind or type.16 Indeed, he claims that

efforts to define the fiduciary relationship rest on “vain hope”.17

Finally, James Edelman claims that it is possible to analyse the ascription of

fiduciary duties independently of an inquiry into the nature of fiduciary relationships.18

He argues that: “we can only understand when fiduciary duties arise if we conceive of

them as obligations based upon manifestations of a voluntary undertaking to another.”19

Edelman considers it unnecessary to characterize the nature of the undertaking, claiming

that fiduciary duties are not “necessarily referable to a relationship or status.”20

B. The Fiduciary Relationship Grounds Fiduciary Liability

The view that the fiduciary relationship is indefinable conflicts with prevailing judicial

opinion, according to which fiduciary liability is premised on breach of one or more

duties occasioned by the formation of a fiduciary relationship. Contra Edelman, fiduciary

duties have historically been “necessarily referable to a relationship.” They arise where a

fiduciary relationship arises and they govern the conduct of fiduciaries within the ambit

of this relationship however far it happens to extend.

15
Matthew Conaglen, The Nature and Function of Fiduciary Loyalty, 121 L.Q.R. 452 (2005).
16
Id. at 454.
17
MATTHEW CONAGLEN, FIDUCIARY LOYALTY: PROTECTING THE DUE PERFORMANCE OF NON-FIDUCIARY
DUTIES 9 (2010).
18
James Edelman, When do Fiduciary Duties Arise? 126 L.Q.R. 302 (2010).
19
Id. at 302.
20
Id.

Electronic copy available at: https://ssrn.com/abstract=2353062

 The conventional wisdom of judges implies the analytical priority of relationship

over duty in the structure of fiduciary liability. This priority is reflected in what judges

say about fiduciary liability as well as what they do in adjudicating liability.

Let us start with what matters most – the practices whereby litigants argue, and

judges adjudicate, fiduciary liability claims. Here, relationship characterization figures

centrally. Plaintiffs do not simply assert a duty and plead facts showing breach. Instead,

they first claim that their relationship with the defendant was fiduciary. More particularly,

they claim that it was fiduciary as a matter of status or fact.21 Courts adjudicate claims

accordingly. They determine whether the relationship was, in fact, fiduciary as a matter

of status or fact. Once that determination is made, recognition of fiduciary duties, and

analysis of allegations of breach, follows.

Appreciation of the role of relationship characterization in fiduciary liability

requires brief explanation of dominant methods of identifying fiduciary relationships.22 In

the usual case, a plaintiff will argue that she and the defendant are in a relationship of

fiduciary status. The relationships that most think of as fiduciary are so recognized as a

matter of status (e.g., agent and principal, trustee and cestui, director and corporation).

Confronted with a claim of this sort, the court will determine whether the relationship is

of a type that has settled fiduciary status. If so, it is almost invariably treated as fiduciary

and the defendant will be deemed to have been subject to fiduciary duties.

In the rare case that a plaintiff is unable to prove a relationship of settled fiduciary

status, she may argue either for a de novo extension of status or a one-off judgment that

the relationship is fiduciary. Sometimes litigants manage to convince the courts that an

21
See D.G. Smith, supra note 8.
22
I discuss these methods in greater detail in Paul B. Miller, A Theory of Fiduciary Liability, 56 MCGILL
L.J. 235 (2011). See also D.G. Smith, supra note 8 and Glover, supra note 5.

Electronic copy available at: https://ssrn.com/abstract=2353062

extension of status is in order. However, the bar is very high given the precedential

implications. It is thus more typically argued that the relationship is fiduciary by virtue of

its actual (rather than presumed) characteristics. Here, facts rather than status drive the

analysis. A relationship is identified as fiduciary by virtue of its possession of

characteristics of recognized fiduciary relationships (i.e., those that have settled fiduciary

status). There is considerable difference of opinion amongst judges on the characteristics

of fiduciary relationships.23 However, commonly cited characteristics include discretion,

power, inequality, vulnerability, trust and confidence.

I have elsewhere criticized these methods of identifying fiduciary relationships.24

However, their defensibility is immaterial here. The important point for present purposes

is that the kind-status of a relationship (whether in its own right or by virtue of its status)

determines whether fiduciary duties apply to it. Fiduciary liability is formally structured

even if we don’t understand well the formal properties of the kind of relationship on

which fiduciary duties are based.

Given that relationship characterization is central to the formulation and

adjudication of liability claims, it should be unsurprising that judges frequently say as

much. Consider Guerin v Canada, in which Dickson CJ had to decide whether the

incidence of fiduciary duties is controlled wholly (and merely) by status or rather follows

from something in the nature of the fiduciary relationship. He found the latter to be the

case:

[I]t is sometimes said that the nature of fiduciary relationships is both established
and exhausted by the standard categories of agent, trustee, partner, director, and

23
Hoyano, supra note 4.
24
Miller, supra note 22.

Electronic copy available at: https://ssrn.com/abstract=2353062

 the like. I do not agree. It is the nature of the relationship, not the specific
category of actor involved that gives rise to the fiduciary duty.25

Also instructive are comments on methods of identifying fiduciary relationships.

For example, in Lac Minerals Ltd v International Corona Resources Ltd, La Forest J said

of the status-based approach that:

The focus is on the identification of relationships which, because of their inherent

purpose or presumed factual or legal incidents, the courts will impose a fiduciary
obligation on one party to act or refrain from acting in a certain way.26

Of the fact-based approach, he said:

The imposition of fiduciary obligations is not limited to those relationships in

which a presumption of such an obligation arises. Rather, a fiduciary obligation
can arise as a matter of fact out of the specific circumstances of a relationship.27

Similar statements are found in other leading cases. For instance, in Hospital

Products Ltd v United States Surgical Corporation, Dawson J made it clear that fiduciary

duties arise from, and are responsive to, properties of the fiduciary relationship:

There is … the notion underlying all cases of fiduciary obligation that inherent in
the nature of the relationship itself is a position of disadvantage or vulnerability
on the part of one of the parties … From that springs the requirement that a
person under a fiduciary obligation shall not put himself in a position where his
interest and duty conflict.28

In the United States, courts have also emphasized that fiduciary duties arise by

virtue of fiduciary relationships. For example, in United States v Chestman, the U.S.

Court of Appeals for the Second Circuit said:

A fiduciary relationship involves discretionary authority and dependency … the

beneficiary of the relation may entrust the fiduciary with custody over property of
one sort or another. Because the fiduciary obtains access to this property to serve

25
[1984] 2 SCR 335 (SCC) 341 (emphasis supplied).
26
[1989] 2 SCR 574 (SCC) 646 (emphasis supplied).
27
Id. at 648 (emphasis supplied).
28
(1984) 156 CLR 41 (HCA) 142 (emphasis supplied).

10

Electronic copy available at: https://ssrn.com/abstract=2353062

 the ends of the fiduciary relationship, he becomes duty-bound not to appropriate
the property for his own use.29

Of course, courts are evidently not of one mind on the nature of the fiduciary

relationship or the link between it and fiduciary duties. However, they are predominantly

of the view that fiduciary relationships ground fiduciary duties and thus fiduciary

liability.

3. Resolving the Conflict:

The Fiduciary Powers Theory of the Fiduciary Relationship

We have established that there are two important, but conflicting, views on fiduciary

relationships. Leading academics hold that the fiduciary relationship is indefinable;

judges treat it as a lynchpin of liability. Scholarly opinion reflects real uncertainty about

the nature of the fiduciary relationship. However, in the conflict between it and the

judicial opinion, the latter must prevail. Lawyers and judges reason and act as though the

fiduciary relationship is a distinctive form or kind of legal relationship. We cannot but

take these practices seriously. Indeed, I suggest that they supply the basis for definitional

reasoning. In previous work I have offered a working definition of the fiduciary

relationship as part of a broader fiduciary powers-based theory of fiduciary liability.30 I

shall now develop the definition more fully, explain its implications, and argue for its

viability and utility.

29
947 F.2d 551, 569 (2d Cir. 1991) (emphasis supplied). See also the RESTATEMENT OF (THIRD) OF TRUSTS
§ 2 cmt. b: “Despite the differences in the legal circumstances and responsibilities of various fiduciaries,
one characteristic is common to all: a person in a fiduciary relationship to another is under a duty to act for
the benefit of the other as to matters within the scope of the relationship”.
30
Supra note 22, at 261-265.

11

Electronic copy available at: https://ssrn.com/abstract=2353062

A. A Definition of the Fiduciary Relationship

I suggest that the fiduciary relationship is a distinctive and coherent kind of legal

relationship that may be defined as follows:

A fiduciary relationship is one in which one party (the fiduciary) exercises

discretionary power over the significant practical interests of another (the
beneficiary).31

The above definition identifies formal properties of the fiduciary relationship.

These properties establish the content and parameters of the relationship as a general kind

of legal relationship; some are definitive properties, others are structural properties. The

definitive properties are kind-defining terms and thus drive the taxonomical function of

the kind or category of relationship picked out by the definition. Kind-defining terms

make it possible to use the definition to sort, categorize and differentiate particulars. The

structural properties identify implications of kind-status for the parties to a relationship

falling within it.

We may first consider the definitive properties of the fiduciary relationship under

the fiduciary powers theory. The most significant of these is, unsurprisingly, the exercise

of power by the fiduciary relative to the beneficiary. It is commonly suggested that

fiduciaries possess power over others.32 However, power is conceptually ambiguous.

31
Id. at 262.
32
United States v. Chestman, supra note 29; Hospital Products Ltd. v. United States Surgical Corporation,
supra note 28; Norberg v. Wynrib [1992] 2 SCR 226 (SCC); and Galambos v. Perez 2009 SCC 48 (SCC).
See also Shepherd, supra note 8; D.G. Smith, supra note 8; Evan Criddle, Fiduciary Administration:
Rethinking Popular Representation in Agency Rulemaking, 88 TEXAS L. REV. 441 (2010); and EVAN FOX-
DECENT, SOVEREIGNTY’S PROMISE: THE STATE AS FIDUCIARY (2012). While many courts analyze the
relationship in terms of power, some conceive of it in terms of trust and confidence. See the judgment of
Millett LJ in Mothew, supra note 7, at 18: “A fiduciary is someone who has undertaken to act for or on
behalf of another in a particular matter in circumstances which give rise to a relationship of trust and
confidence.” In my view, the stipulation of an undertaking as one to “act for or on behalf of another”
implies a relationship of power of the sort described in Section 3.A, below. This view is resonant with
Mason J’s statement in Hospital Products, id. at 96-97, that “[t]he critical feature of [fiduciary]
relationships is that the fiduciary undertakes or agrees to act for or on behalf of or in the interests of another

12

Electronic copy available at: https://ssrn.com/abstract=2353062

Power may mean control, authority, strength, or influence, amongst other things. And

many of these concepts are, in turn, ambiguous. Even Hohfeld’s well-known definition of

a legal power won’t do.33 Hohfeld conceives of legal powers in general as capacities

(devised or sanctioned by law) to alter the legal position or relationship of another.34

Whether or not this is an adequate account of the juridical character of legal powers in

general, it does not fully encompass fiduciary powers or express well their juridical

character. Some fiduciaries in some cases wield powers over the material rather than

legal interests of beneficiaries (e.g., physicians and parents) and in exercising those

powers influence the material but not legal position of the beneficiary. The possession of

such a power alters the terms on which the fiduciary and beneficiary and/or benefactor of

the power relate (inasmuch as possession of these powers entails authorization in the

sense described below), but the exercise of the power needn’t alter the legal character of

these or other relationships much less the legal position of the parties to them. Fiduciaries

in these relationships wield a kind of legal power not captured by Hohfeld’s definition.

Similarly, some people wield powers that are clearly legal powers in Hohfeld’s sense but

are not powers that could not reasonably be considered fiduciary. The simplest example

may be the exercise of powers inherent in options arising under contract (e.g., an option

to purchase land at an agreed price within a fixed term). The exercise of an option

influences the legal position or relationship of another (at the very least, the counter-party

who gave the option and is liable to its exercise) but the power to act on an option is not

person in the exercise of a power or discretion which will affect the interests of that other person in a legal
or practical sense” (emphasis supplied).
33
Compare Lionel Smith, Can We Be Obliged to Be Selfless? in this volume.
34
See Wesley N. Hohfeld, Some Fundamental Legal Conceptions as Applied in Judicial Reasoning, 23
YALE L.J. 16, 55 (1913): “a power is one’s affirmative ‘control’ over a given legal relation as against
another.”

13

Electronic copy available at: https://ssrn.com/abstract=2353062

(or is at least not inherently35) other-regarding; instead, it may be and ordinarily is

exercised self-interestedly. To say simply that a fiduciary relationship is one in which one

person possesses power – or even legal power of some sort – relative to another is

therefore inadequate. The concept of fiduciary power must be disambiguated from other

varieties of power.

I suggest that fiduciary power is distinguishable from other varieties of power by

virtue of the fact that it is a form of authority ordinarily derived from the legal

personality of another (natural or artificial) person.36 The fiduciary, by virtue of the

power vested in her, stands in substitution for the beneficiary or a benefactor in

exercising a legal capacity that is ordinarily derived from the beneficiary or benefactor’s

legal personality.37 Personal legal capacity consists in the ability to act in legally effective

35
I recognize that an option may be held under a fiduciary mandate in which case its exercise of is other-
regarding in the sense that the fiduciary is obligated to act in the interests of the beneficiary. But this is a
function of the fact that exercise of the option (a kind of legal power) is subject to the exercise of fiduciary
power (a different kind of legal power). There is nothing in the juridical character of the option itself that
makes it a kind of power that attracts fiduciary duties.
36
I therefore follow Hobbes in conceiving of fiduciary authority as inherently representational and rooted
in the concept of personality as such; persons acting on those capacities that are inherent in their legal
personality and held relative to their person or property needn’t claim authority as such for the rightfulness
of their action is entailed by the connection between the capacities and their personality. Persons may be
said to act with authority when they exercise these powers or capacities but in speaking of authority as such
we are ordinarily concerned with their conferral – i.e., with the authorization upon which legal
representation depends. Authority as a form of representation signifies that the exercise of conferred (i.e.,
acquired) powers is rightful. The person wielding fiduciary authority invariably does so as a representative;
it is for that reason that Hobbes speaks of those acting with authority on behalf of another as “personating”
the other. Acting with authority in this sense is an extension, through representation, of the personality of
another (or of oneself as donor of powers given over to another). See THOMAS HOBBES, LEVIATHAN 112
(R. Tuck ed. 1996): “the Right of doing any Action, is called AUTHORITY and sometimes warrant. So
that by Authority, is alwayes understood a Right of doing any act; and done by Authority, done by
Commission, or Licence from him whose right it is.” Hobbes’ account of authority is distinctive in its
capacity to encompass private and public forms of fiduciary power. On the significance of the concept to
our understanding of the state, and its exercise of public powers, as fiduciary see FOX-DECENT, supra note
32, at 5-13. See also DAVID RUNCIMAN, PLURALISM AND THE PERSONALITY OF THE STATE 6-33 (1997).
37
I repeat with the caveat that the power is “ordinarily” derivative in recognition of the fact that in a limited
set of circumstances a fiduciary may declare that a capacity of her own will be henceforth held by her in a
fiduciary basis for the benefit of a beneficiary. The transformation of personal powers to fiduciary powers
by declaration may be part and parcel of a benefaction (as is true of self-declaratory trusts) but it may also
arise in the context of incorporation or partnership (as where a director or partner transfers personal
property to a corporation or partnership). In such cases the representative character of fiduciary power

14

Electronic copy available at: https://ssrn.com/abstract=2353062

ways as well as the susceptibility to being held responsible for so acting. It is through the

exercise of legal capacities that one acts as a legal person as such.38

Given that fiduciary power consists in the substitutive exercise of legal capacity,

it follows that varieties of fiduciary power reflect varieties of legal capacity. Fiduciary

power may thus entail the capacity to: enter into legally binding relationships for another,

including contractual relationships; acquire, invest, use, administer or alienate property

owned by or held for another; to authorize the use of another’s name, likeness, or

confidential information; to license or permit access to, contact with, or use of, another’s

physical person or property; to make decisions relating to the health and personal welfare

of another; to hold and claim benefit entitlements; to institute legal proceedings to

enforce or seek vindication of one’s legal rights for another; and to be held responsible

for or answer claims of civil, criminal or regulatory liability of another.39

Fiduciary power as a form of authority implies the freedom to engage in conduct,

and more specifically to make decisions, not otherwise open to its bearer.40 This follows

reflects not its derivation as such but the fact that, by virtue of an underlying benefaction or transaction, the
power has been devoted to an object and as such no longer belongs within the residuum of powers and
capacities through which the donor may pursue ends of his choice. The fiduciary who declares herself to be
a fiduciary in respect of the disposition of certain powers (and, normally, associated property) stands in
substitution for her beneficiary in the sense that her now-fiduciary powers have been dedicated to the ends
of her chosen beneficiary. She exercises them for, or on behalf of, her beneficiary rather than on her own
behalf. On the substitutive or representational character of fiduciary power, see the judgment of Mason J. in
Hospital Products, supra note 28, at 97: “The expressions ‘for’, ‘on behalf of’ and ‘in the interests of’
signify that the fiduciary acts in a ‘representative’ character.”
38 Exercise of legal capacity is ordinarily, but not inherently, volitional; sometimes a person will be taken to

have acted in a legally effective way notwithstanding that he did not specifically intend to so act (let alone
to bring about associated legal consequences).
39
This list is illustrative, not exhaustive.
40
This should be uncontroversial in respect of most fiduciaries; however, is true even where it seems
nonsensical. One might think that a fiduciary who has declared that certain of her personal powers will be
held on a fiduciary basis is not, in subsequently exercising those powers in a fiduciary capacity, doing
anything other than what she may have freely done in exercising the powers in a personal capacity. But this
thought is mistaken. It fails to account for the normative implications of the benefaction or transaction
through which the powers were devoted to the beneficiary. Whether the transformation in the character of
the power is accomplished by way of gift or through contract, the powers must, from the moment at which
fiduciary mandate has been formalized, be taken as powers belonging rightfully to the beneficiary (the

15

Electronic copy available at: https://ssrn.com/abstract=2353062

from the fact that capacities associated with legal personality are themselves personal in

nature. The individual possessed of personality is the presumptive bearer of the legal

capacity to act in her own name and to be recognized and held accountable for so acting.

Thus the capacity to enter into a contact is presumptively held by individuals in their own

right rather than by agents (agency requires showing of authority). Likewise, the capacity

to decide whether to consent to treatment presumptively lies in the patient who will be

subject to it (substitute decision-making likewise requiring proof of authority). The

authorization through which the fiduciary is invested with fiduciary powers has a

transformative effect. It alters the normative basis upon which fiduciaries, beneficiaries,

and benefactors interact for it enables (literally, authorizes) the fiduciary to act in ways

she would otherwise be legally incapable of and/or prohibited from acting.

In some cases, the substitutive exercise of legal capacity might simply entail an

extension of legal personality through which the capacity is given effect by another. For

example, one person might cede a capacity to another but dictate how it is to be

exercised. In such cases, the legal and practical effect of the substitution is limited. The

person exercising the capacity simply effectuates the will of the person from whom it was

derived. The law does facilitate substitutive exercise of capacity in this rather thin sense.

Nevertheless, these powers are normally not considered fiduciary. As prominent

authorities attest, powers are ordinarily considered fiduciary only if they are

discretionary.41 Discretion entails freedom of choice in the exercise of the power.42 This

cestui, the partnership, or the corporation, as the case may be). The fiduciary cannot rightfully exercise
those powers as a fiduciary unless she has been properly constituted a fiduciary (i.e., unless her acts have
effectively authorized her to carry on in the exercise of the transferred powers in her new, fiduciary,
capacity).
41
See United States v. Chestman, supra note 29, at 569: “A fiduciary relationship involves discretionary
authority and dependency”; Galambos v. Perez, supra note 32, at para 70: “The particular relationships on
which fiduciary law focuses are those in which one party is given a discretionary power to affect the legal

16

Electronic copy available at: https://ssrn.com/abstract=2353062

in turn means that beneficiaries and benefactors are not merely subject to the conduct of

fiduciaries; they are subject as well to their will. Fiduciaries decide, as well as act, for

others.

The requirement of discretion is important but should not be overstated. The

fiduciary has freedom of choice and action but it is not unlimited. Indeed, it is also an

essential characteristic of fiduciary power that it is specified. Specification is ordinarily

found in the express terms of a grant of authority but may also be supplied by law.

Fiduciary power may be specified in any number of ways. It may be subject to triggering

conditions (i.e., powers may arise at a certain date or upon the occurrence of a certain

event). It will usually be subject to conditions that define its scope (i.e., specific matters

or mandates are reserved for the judgment of the fiduciary) or constrain its exercise (e.g.,

considerations are mandated or prioritized).

Fiduciary power is also inherently relational. Private fiduciaries are not

sovereigns; they do not enjoy authority at large relative to a populace with diffuse and

markedly divergent interests. Rather, their authority is held relative to particular

beneficiaries and benefactors with clearly defined personal or common interests.

Fiduciaries exercise power in the world at large and thus relative to third parties but they

hold it relative to a specific individual or group. The relational character of fiduciary

power is a function of its derivation and implied purpose. Fiduciary power is derived

from the legal personality of another and/or is expressly devoted to the ends of another;

or vital practical interests of another”; and Weinrib, supra note 8, at 5: “if [the fiduciary has] no discretion
to advise or negotiate and if their instructions are narrow and precise there is nothing on which the
fiduciary obligation can bite.” See also authorities cited in Miller, supra note 22, at 261-263 and D.G.
Smith, supra note 8, at 1397.
42
On the relationship between discretion in contractual performance and the execution of fiduciary
mandates, see D. Gordon Smith and Jordan Lee, Discretion, 78 OHIO STATE L.J. (forthcoming).

17

Electronic copy available at: https://ssrn.com/abstract=2353062

in any case, it founds a relationship in which one acts for the purpose of advancing the

ends of another.

The relational character of fiduciary power provides a natural segue to another

definitive property of the fiduciary relationship; namely, that practical interests of the

beneficiary serve as the ground of fiduciary power. Practical interests include matters of

personality, welfare, or right pertaining to persons or their causes. Matters of personality

include determination of the interests of incapable persons. Matters of welfare include

decisions pertinent to the physical, mental and material well-being of persons or to the

advancement of their causes. Matters of right include decisions relating to the exercise,

enforcement, performance, discharge or alienation of legal rights, obligations, powers,

and liabilities of persons. Practical interests are, as a matter of conceptual necessity, such

that they can be implicated by an individual acting personally on one or more of her legal

capacities. The requirement of significance follows. Significance is not a function of the

value or importance of an interest. It is, rather, a matter of the objective subsistence of an

interest, such that the law can recognize it as such. An interest is not significant in this

sense if a casual observer cannot see how it is or might be affected by the exercise of a

legal capacity.

Structural formal properties of the fiduciary relationship are, by their nature, not

immediately apparent in the above definition of the relationship. However, they, too,

reflect the nature of fiduciary power and thus form an important part of the fiduciary

powers theory of the fiduciary relationship. There are three related structural properties of

the fiduciary relationship. The first is inequality. The fiduciary is, simply by virtue of the

establishment of a fiduciary relationship, in a dominant position relative to the

18

Electronic copy available at: https://ssrn.com/abstract=2353062

beneficiary. Wherever one person enjoys discretionary power over the practical interests

of another, their relationship will be unequal in respect of the possession and exercise of

fiduciary power itself. It must be emphasized that this inequality, being structural, is a

matter independent of any circumstantial inequality that might exist between the parties.

The beneficiary is always subordinate in respect of the possession and exercise of

fiduciary power itself, notwithstanding that in other respects he may be ascendant.

The second and third structural properties of the fiduciary relationship are

dependence and vulnerability. These properties are each, in turn, reflections of the

structural inequality generated by the formation of a fiduciary relationship. Fiduciary

power entails influence, including the risk of adverse influence. The fiduciary is to

exercise power to achieve certain ends for the beneficiary. The beneficiary is thus

invariably dependent upon the fiduciary as power is exercised so as to affect her practical

interests. Vulnerability is a corollary of dependence, so understood. Fiduciaries are

expected to exercise power to salutatory effect, but beneficiaries’ interests might instead

be set back. The degree of vulnerability faced by the beneficiary might be controlled in

various ways (e.g., by incentive, monitoring and enforcement mechanisms). However, it

cannot be eliminated. The susceptibility of the beneficiary to the influence of the

fiduciary in the exercise of fiduciary power entails some risk of adverse influence.

B. Some Doctrinal Implications

The fiduciary powers theory of the fiduciary relationship has implications for key

doctrinal issues. I can mention only a few presently and so will briefly comment on

19

Electronic copy available at: https://ssrn.com/abstract=2353062

implications relating to the identification of fiduciary relationships, the incidence of

fiduciary duties, and the scope of fiduciary duties.

The most important implications of the fiduciary powers theory arise simply by

virtue of its clarification of the formal properties of the fiduciary relationship. As a

general level, the fiduciary powers theory contributes to broader efforts to map the

conceptual landscape of private law.43 It does so by clarifying the nature of the fiduciary

relationship such that it may be distinguished from other kinds of private law

relationship, and by clarifying the incidence and function of fiduciary duties relative to

other kinds of private law duty. More significant, though, is the primary sorting function

served by the definition of the fiduciary relationship in the fiduciary powers theory.

Existing methods of identifying fiduciary relationships do not perform this function well

because they sort by approximation.44 By contrast, a definition of the sort offered here

permits sorting directly by reference to properties of the kind or category on which the

whole enterprise of sorting is premised. Faced with a particular relationship, one can

simply determine whether it possesses properties constitutive of the kind or category of

relationship and so falls within or outside it.

Another important issue is that of determining the incidence of fiduciary duties.

When do fiduciary duties arise? This is, as Edelman notes, a matter of some

uncertainty.45 The dominant judicial view supplies a pat answer: When a fiduciary

relationship is established. This answer is important but has been unsatisfying given

prevailing confusion over the nature of the fiduciary relationship. If it is unclear what

43
PETER BIRKS, ENGLISH PRIVATE LAW (2004) and ANDREW BURROWS AND LORD RODGER OF
EARLSFERRY, EDS., MAPPING THE LAW: ESSAYS IN MEMORY OF PETER BIRKS (2006).
44
See Miller, supra note 22, at 247-252.
45
See Edelman, supra note 18.

20

Electronic copy available at: https://ssrn.com/abstract=2353062

makes a relationship fiduciary, it must be unclear when and how a fiduciary relationship

may be established and, by extension, when fiduciary duties will arise. The fiduciary

powers theory offers clarity. Fiduciary duties arise where one person receives or

undertakes discretionary power over the significant practical interests of another. The

nature of the relationship determines the means by which it may be formed; these

methods of relationship formation amount to modes of authorization.

There are three principal modes of authorization: mutual consent, unilateral

undertaking, and legal decree. In most cases, given the nature of the power conveyed, the

relationship is established by mutual consent. Consent is often signified by contract. But

it may be demonstrated otherwise (e.g., by written consent, oral agreement, or by

implication). However evidenced, mutual consent founds a fiduciary relationship by

signaling that transferor and transferee intend the underlying transfer of power. Where

the transfer of power is part of a benefaction it is impressed with the donative intent of

the benefaction itself. More rarely, a fiduciary relationship may be established through

unilateral undertaking by the fiduciary (e.g., voluntary assumption of fiduciary power) or

by decree (e.g., through legislation or by court order). Usually, recourse to these modes

of authorization is made necessary by the incapacity of the beneficiary or benefactor to

consent or by the impracticality of authorization being provided on a case by case basis.46

A final issue is that of determining the scope of fiduciary duties.47 This issue has

proven particularly pressing in commercial fiduciary relationships. In such relationships,

the fiduciary will often have a pecuniary interest in an area of business in which her

46
Though in the case of personal powers that are declared fiduciary unilateral undertaking is the only
appropriate mode of authorization; the beneficiary has no standing to consent or object to the authorization
as such.
47
Lionel Smith, The Motive, Not the Deed, in RATIONALIZING PROPERTY, EQUITY AND TRUSTS 53 (J.
Getzler ed., LexisNexis 2003).

21

Electronic copy available at: https://ssrn.com/abstract=2353062

beneficiary has, or might come to have, a like interest. It is virtually impossible to rule

out potential conflicts. This has become apparent in cases involving alleged appropriation

of business opportunities.48 A disappointed beneficiary has an incentive to over-state the

scope of fiduciary duty to reach profits realized on an opportunity by a fiduciary. Courts

have recognized the importance of delineating the scope of fiduciary duties, but have

struggled to find a principled basis for, and method of, delineation.

The fiduciary powers theory fills that gap. Under this theory, the fiduciary

relationship grounds fiduciary duties and determines their sphere of operation. It suggests

that fiduciary duties ensure that fiduciary powers are exercised in a manner consistent

with the beneficiary’s exclusive claim relative to them. Fiduciary duties have no juridical

purpose independent of the relationship that gives rise to them. It follows that, to define

the scope of fiduciary duties, one must define the scope of the fiduciary relationship. This

is necessarily a fact-based exercise in construction of the terms of particular fiduciary

mandates. However, attention to the formal character of the fiduciary relationship allows

one to focus the construction.

To ascertain the scope of a fiduciary relationship, and by extension the sphere of

operation of fiduciary duties, it is necessary that one clarify the relative positioning of

fiduciary and beneficiary under the mandate in question. One must identify particular

capacities held by the fiduciary and the practical interest(s) of the beneficiary directly

implicated by their exercise. One must also determine whether particular purposes have

been specified (or may be implied) in the grant of power. In this way, the scope of

fiduciary duties may be ascertained by determining what the fiduciary is expected to

48
See, for example, Peso Silver Mines Ltd (NPL) v. Cropper [1966] SCR 673 (SCC); Industrial
Development Consultants v. Cooley [1972] 1 WLR 443; Canadian Aero Service Ltd v. O’Malley [1974]
SCR 592 (SCC); and Bhullar v. Bhullar [2003] EWCA Civ 424.

22

Electronic copy available at: https://ssrn.com/abstract=2353062

accomplish for the beneficiary through the exercise of which powers. Fiduciary duties

constrain the conduct of the fiduciary within the ambit of the relationship so ascertained,

but not beyond it.

4. Testing the Fiduciary Powers Theory: Core Cases

The fiduciary powers theory of the fiduciary relationship suggests that identification of

fiduciary relationships should proceed by way of definitional reasoning. A critical test of

any definition of the fiduciary relationship lies in its ability to encompass relationships of

unquestioned fiduciary status. Status is an imperfect proxy for definitional reasoning, but

we should have little reason for confidence in a definition of a concept if it does not

comport with common usage of the concept. In fiduciary law, we have greatest apparent

confidence in the fiduciary nature of a few core categories of relationship. In this section,

I will show how the fiduciary powers theory of the fiduciary relationship explains the

presumptive fiduciary status of trustees, agents, and corporate directors.

A. Trustees as Fiduciaries

The trustee is widely considered to be the prototypical fiduciary.49 When asked what

makes the trustee a fiduciary, some hearken to the Latinate fiducia and its signification of

social practices of trust reposed and undertaken in respect of property.50 It is unclear

whether this is as insightful as is supposed (mostly because it is unclear what it means to

repose and undertake trust in general), but even supposing that trusteeship reflects a

49
By “trustee” I here mean the trustee of an ordinary express trust with a primarily donative purpose. I shall
not presently explore complexities associated with public trusts, commercial trusts, and remedial trusts.
50
Birks, supra note 6.

23

Electronic copy available at: https://ssrn.com/abstract=2353062

social economy of trust it is also the case that a trustee is one in whom legal power is

reposed.

The notion that the trustee is a fiduciary by virtue of his possession of legal

powers over trust property is not new.51 Nevertheless, the basis for the fiduciary

characterization of these powers has rarely been explained. I suggest that ordinary powers

of trusteeship (administrative and dispositive) are fiduciary powers in the sense explained

in Section 3.A. The trustee of a private donative trust acts as a trustee through the

authority he enjoys as such. This authority is in the nature of legal capacity to act in

legally effective ways relative to the trust corpus. The powers of the trustee are

sometimes (perhaps misleadingly) said to be those entailed by legal ownership of trust

property. Precise delineation of these powers is a matter best left to treatise writers. But

even cursory analysis reveals that trustees exercise legal capacities ordinarily associated

with legal ownership. These capacities include powers of appointment, distribution,

51
Many authorities make reference to these powers when defining the nature of the trust or the office of
trusteeship. Several statutes likewise refer to offices of trusteeship in terms of administrative and
investment powers in relation to trust property. See, for example, HALSBURY’S LAWS OF ENGLAND, VOL.
25 at 5: “A trust, in the modern and confined sense of the word, is a confidence reposed in a person with
respect to property, of which he has possession or over which he can exercise a power, to the intent that he
may hold the property or exercise the power for the benefit of some other person or object”; and the reasons
of Lord Eldon in Brown v. Higgs (1801) 8 Ves. Jun. 561, 570, 574: “[T]here are not only a mere trust and a
mere power, but there is also known to the Court a power which the party to whom it is given is intrusted
and required to execute; and with regard to that species of power, the Court considers it as partaking so
much of the nature and qualities of a trust, that if the person, who has that duty imposed on him does not
discharge it, the Court will to a certain extent discharge the duty in his room and place.” See also
RESTATEMENT (SECOND) OF TRUSTS, Vol. 1, Ch. 7.3 – Powers of the Trustee; Trustee Act, 1925, 15 Geo. 5
Ch. 19 (describing general and particular powers of trustees); Trustee Act, R.S.O. 1990, c. T23 (likewise,
characterizing trusteeship in terms of possession of powers); J.E. PENNER, THE LAW OF TRUSTS 21 (6th ed.
2008), explaining that, by virtue of the division of legal and equitable title, “the trustee has all of the legal
rights and powers associated with the property” and explaining the nature of the trust and relationships
between its constituents in terms of powers and constraints thereon; and GRAHAM VIRGO, THE PRINCIPLES
OF EQUITY AND TRUSTS 381 (2012), explaining that: “once appointed, trustees have a wide variety of
powers” and explaining the nature of trusteeship in terms of possession of powers subject to fiduciary and
other constraints. While some have endeavored to distinguish trusts and powers of trusteeship, courts have
considered the concept of power to be so essential to the nature of the trust as to make any such distinction
artificial. See McPhail v. Doulton [1971] A.C. 242 (HL) 448G per Lord Wilberforce “It is striking how
narrow and in a sense artificial is the distinction...between trusts, or as the particular type of trust is called,
trust powers and powers...A layman and, I suspect, a logician would find it hard to understand what the
difference is.”

24

Electronic copy available at: https://ssrn.com/abstract=2353062

maintenance, and advancement.52 They also include powers of delegation and of

satisfying debts and settling claims.53 Needless to say, only one with powers of

ownership - held personally or on trust - is permitted to make decisions about the

investment or maintenance of property, to license access to or use of the property, and so

on.

The powers wielded by trustees are derivative, discretionary, relational, and

specified within the meaning of the fiduciary powers theory. As is true of most

fiduciaries, most trustees enjoy their powers derivatively. Apart from trustees acting

under self-declaratory trusts, they obtain their powers through authorization by another.

In some cases (e.g., where trust property is that of a person who died intestate), trustees

obtain their authority by decree (e.g., under a Trustees Act). However, in most cases it is

received from a settlor via a trust deed. An ordinary express trust for a donative purpose

effectuates a benefaction of property and power for the benefit of another. The power and

property are both derived from the settlor, who in settling the trust gives effect to her

intent to do more than merely pass title by gift.

The powers of a trustee are specified. Indeed, specificity is a condition of validity

for express trusts, as reflected in the ‘three certainties’ requirements.54 Amongst other

things, for a purported declaration of trust to be effective in creating a trust, the powers

devolved upon the trustee must be certain in their own terms as well as in respect of their

objects (purposes or beneficiaries) and the property to which they relate. Beyond these de

minimis requirements, the powers of trustees are ordinarily further specified in the trust

deed.

52
See VIRGO, id. at 459-478.
53
Id. at 454-457 and 458.
54
Knight v. Knight (1840) 3 Beav 148.

25

Electronic copy available at: https://ssrn.com/abstract=2353062

 The discretionary nature of trustees’ powers is reflected in the traditional

distinction between bare versus fixed and discretionary trusts. Under bare trusts, the

trustee has no discretion in the administration of the trust. He serves as mere nominee,

taking title but acting at the direction of the settlor-beneficiary. Bare trustees have passive

responsibilities of maintenance but no positive fiduciary mandate requiring exercise of

independent judgment. It is for this reason that bare trustees are not fiduciaries.55 The

trustee of a fixed trust, in contrast to that of a discretionary trust, does not have discretion

in determining the interests of beneficiaries in trust property. But trustees of both fixed

and discretionary trusts have discretion in administering the trust (e.g., in making

investment and maintenance decisions). It is thus unsurprising that no one doubts the

fiduciary character of trusteeship of fixed and discretionary trusts.

The powers of trustees are also relational. However the trust is effectuated, the

trustee receives her authority through an intentional transfer of property and power.

Powers are settled on trust along with property because this is essential to the settlor

achieving her purpose for the transfer – i.e., making a mediated, and in most cases

conditional, benefaction. As is true generally, authorization has important relational

effects. Whereas in most fiduciary relationships, it alters the legal basis on which

fiduciary and beneficiary interact, with trusts things are rather more complicated. Here,

55
See Ironside v. Smith, 1998 ABCA 366 (per Fruman J: “An individual may hold property on behalf of
another as bare trustee without taking on all the onerous trappings of a fiduciary. A bare trustee has no
further duty to perform except to convey the property to the beneficiary on demand and, so long as he holds
it, to exercise reasonable care over the property, by maintaining or investing it”) and Financial
Management Inc. v. Associated Financial Planners Ltd. (2006) 367 W.A.C. 70 (per McFayden, Picard and
O’Brien JJ: “while a trustee may owe fiduciary duties to its beneficiary, a bare trustee does not.”). See also
PENNER, supra note 51, at 435-436, saying, of bare trusts, “Recall that for fiduciary obligations to arise,
there must be some scope for discretion or leeway in the fiduciary’s performance of her duties. No such
discretion or leeway arises in the case of nomineeship, because in such cases the trustee is only to follow
the beneficiary’s instructions exactly. If the nominee breaches the trust in some way, he will certainly be
liable for breach of trust, but as we know, not all breaches of trust are breaches of fiduciary obligation.”

26

Electronic copy available at: https://ssrn.com/abstract=2353062

the authorization transforms the terms on which settlor, fiduciary and cestui relate, and it

does so in different ways depending on the nature of the trust.

In self-declaratory trusts, the declaration entails transformation in the character of

powers held with respect to the property settled on trust; these powers are, by virtue of

the declaration, henceforth to be held by the settlor-trustee for the benefit of the cestui,

bringing the parties into a new legal relationship with respect to the property and its

administration. The settlor-trustee, in settling the trust, renders fiduciary powers that

would otherwise be personal in the course of making a conditional transfer of property;

the cestui receives, in consequence, new rights relative to the settlor-trustee in respect of

both the exercise of powers and the disposition of associated property.

In ordinary express trusts, a valid declaration of trust effectuates a transfer of

power from settlor to trustee; the trustee enjoys power over property formerly held by the

settlor as legal owner. But given that a private donative trust is, first and foremost, a

benefaction, the most significant relational effect is that which obtains between trustee

and cestui. An ordinary express trust is a means by which to make a mediated benefaction

that might otherwise have been unmediated. In establishing a trust rather than making a

gift, the settlor interposes the trustee between herself and her intended beneficiaries. An

implication of the interpolation of the trustee is that the beneficial interest of cestui in the

trust is made subject to the authority of the trustee. Depending on the terms of the trust, it

may be for the trustee to determine whether, when, how, and how far the beneficial

interest of the cestui is to be served in the maintenance, investment, disposition and

distribution of trust property.

27

Electronic copy available at: https://ssrn.com/abstract=2353062

 It is sometimes said that one of the defining features of the trust is that trustees

receive rights or powers relative to property subject to duties to exercise those rights or

powers in the interests of the cestui.56 We are now in a position to better appreciate the

truth of statements like this. Trustees are fiduciaries, and so subject to fiduciary duties,

simply by virtue of the nature of the powers under which they act as trustees.

B. Agents as Fiduciaries

The presumptively fiduciary character of agency is a matter of general consensus as well.

Like trustees, agents enjoy power relative to their beneficiaries.57 These powers are also

evidently fiduciary.

Agency powers are powers in the sense described in Section 3.A; namely, they

are legal capacities derived from the legal personality of another person. In the context of

agency, that person is the principal. An agent in the technical legal sense is one with

authority to make legally binding arrangements for her principal.58 This authority takes

56
F.W. MAITLAND, EQUITY 44 (1936): “I should define a trust in some such way as the following, - when
a person has rights which he is bound to exercise on behalf of another or for the accomplishment of some
particular purpose, he is said to have those rights in trust for that other or for that purpose and he is called a
trustee”; UNDERHILL ON TRUSTS 1 (2010): “A trust is an equitable obligation, either expressly undertaken,
or constructively imposed by the court, under which the obligor (who is called a trustee) is bound to deal
with certain property over which he has control (and which is called trust property), for the benefit of
certain persons (who are called the beneficiaries or cestuis que trust)”; and Lionel Smith, Trust and
Patrimony (2008) 38 REV. GEN. DROIT 379, 381 (2008): “the trust beneficiaries hold rights in the rights
that the trustee holds as trust property” (original emphasis).
57
RESTATEMENT (THIRD) OF AGENCY, §1.01, Comment c.: “As defined by the common law, the concept of
agency posits a consensual relationship in which one person … acts as a representative of or otherwise acts
on behalf of another with power to affect the legal rights and duties of the other person … The common-
law definition requires that an agent hold power; MECHAM ON AGENCY sec. 26 (2nd ed. 1914): ““An agent”
is a person who has authority, express or implied, to act on behalf of another person (the “principal”), and
to bind that other person by his acts or defaults.”; Warren A. Seavey, The Rationale of Agency, 29 YALE
L.J. 859, 861 (1920): “a principal is bound where the agent acts in the exercise of power” (but insisting
upon a sharp distinction between power and authority); Deborah A. DeMott, Disloyal Agents, 58 ALA. L.
REV. 1049, 1050-1051 (2007), explaining that the language “on behalf of” in standard definitions of agency
means “that an agent acts with power to affect the principal’s legal relations.”
58
I follow DeMott in distinguishing the legal conception of agency from colloquial and other (e.g.,
economic) conceptions. Id.

28

Electronic copy available at: https://ssrn.com/abstract=2353062

the form of specific legal capacities that are presumptively personal capacities of the

principal. Consider the most typical kind of agency power – the power to enter into, and

act under, a legally valid contract on behalf of the principal. The power to contract is a

legal capacity inherent in the legal personality of one with actual capacity to act on it (an

adult of sound mind). Conferral of this power on an agent brings about a substitution. The

agent stands in for the principal in exercising powers to make, perform, breach or

renounce contracts that will be binding just as if these powers had been exercised

personally. Consider another kind of agency power – the power of a lawyer to make

binding representations on behalf of her client in filing, answering, amending, and

defending claims, in drafting and filing legal documents, in making submissions to

regulators, or in negotiating transactions.59 The agency powers exercised by lawyers are

also legal capacities derived from the legal personality of the person on whose behalf

they are exercised (i.e., the client). People may, and sometimes against strenuous advice

do, exercise these capacities personally. But they typically cede them to a lawyer. In

doing so, they authorize the lawyer to act in their stead. In accepting a retainer and

undertaking powers associated with it, the lawyer acts as a fiduciary of her client.

Agency powers are also derivative, discretionary, relational, and specified within

the meaning of the fiduciary powers theory of the fiduciary relationship. Their derivative

character should be plain. The agent does not enjoy her powers innately. No one has the

power, at large, to make legally binding arrangements for another. Instead, agency

powers are derived from the legal personality of the principal. The agent cannot act as

59
See generally Deborah A. DeMott, The Lawyer as Agent, 67 FORDHAM L. REV. 301 (1998).

29

Electronic copy available at: https://ssrn.com/abstract=2353062

such (i.e., as an agent) without authorization because her power is derived wholly from

the authorization provided by the principal.60

The specificity of agency powers is amply evidenced in practice. Contracts,

retainers, and the like specify agency powers in a variety of ways. The power(s)

themselves will ordinarily be described (e.g., a power to enter into a contract of a

particular sort). The object of the power will usually also be specified (e.g., a power to

renegotiate a contract of employment). A temporal term will typically be placed on the

agent’s powers. Finally, the exercise of agency powers is characteristically conditioned

by instructions from the principal.61

Unlike trustees, agents act subject to the instruction of their beneficiaries.

However, virtually all agents have discretion in the exercise of powers notwithstanding

the presence of instructions.62 This may be so even where instructions are quite detailed,

insofar as they are subject to interpretation by the agent. Consider the purchasing agent

who acts under instructions to obtain a particular good for a specified maximum price,

but is otherwise free to negotiate purchase terms. The agent’s discretion is limited by ex

ante instruction and may be further limited by ex post consultation, but very rarely is it

eliminated. This is unsurprising, for usually it would be pointless to hire an agent without

60
See RESTATEMENT (THIRD) OF AGENCY §1.01, Comment c: “An agent who has actual authority holds
power as a result of a voluntary conferral by the principal and is privileged, in relation to the principal, to
exercise that power.”
61
See DeMott, supra note 57, at 1051: “A defining characteristic of relationships that are ones of agency
under the common law is the principal’s power to give interim instructions to the agent although principal
and agent have previously agreed the principal will not give such instructions, and although, in giving
them, the principal breaches a contract with the agent. This power is a crucial aspect of the principal’s
position of control over the agent, itself necessitated by the agent’s power to subject the principal to
liability to third parties.”
62
The personal character of fiduciary liability and its dependence on there being an element of discretion in
the exercise of a power conferred is reflected, even if only indirectly, in the rule prohibiting delegation by
agents where the agent’s powers are discretionary. Summers v. Commercial Union Assurance Co. (1881), 6
S.C.R. 19 (S.C.C.)

30

Electronic copy available at: https://ssrn.com/abstract=2353062

affording her some discretion. If the principal intends not to avail himself of her

judgment, he would do as well to exercise the powers personally.

Finally, agency powers are relational. An agent has the authority to act for and on

behalf of the principal in dealing with third parties. The authorization of an agent

transforms the basis on which these parties interact. The principal will be held

accountable for actions of the agent in the exercise of the specific powers vested in the

agent as if the actions were her own. The agent, otherwise incapable of binding the

principal, obtains that ability. Possession and exercise of agency powers also alters the

basis on which agent and principal interact with third parties. Provided notice is given of

the agency relationship, contracts or other transactions entered into by an agent for a

principal with a third party will be binding on the principal rather than the agent. The

direct, personal, and bilateral character of most ordinary legal arrangements is thereby

altered by agency.

C. Directors as Fiduciaries

Corporate directors are quintessential fiduciaries.63 And they, too, are commonly said to

be fiduciaries insofar as they occupy a position of power.64 Indeed, the fiduciary

administration of corporations was famously long ago analogized with that of trusts on

63
Justice J. Walsh, The Fiduciary Foundation of Corporate Law, 27 J. CORP. L 333 (2002) and Leo Strine
et al., Loyalty’s Core Demand: The Defining Role of Good Faith in Corporation Law, 98 GEO. L.J. 629
(2010).
64
See Strine, id. at 633 “Delaware law has traditionally subjected … the use of authority under the statutory
corporate law to the important condition that fiduciary power be exercised for proper corporate reasons and
not to advance a personal agenda of any kind”; and Margaret Blair and Lynn Stout, A Team Production
Theory of Corporate Law, 85 VA. L. REV. 247, 291 (1999): “Like trustees, directors, once elected, become
the ultimate decision-making authority within the firm, constrained primarily by their fiduciary duties.”

31

Electronic copy available at: https://ssrn.com/abstract=2353062

the basis that directors, like trustees, wield power over property for the benefit of

others.65

As we can now appreciate, the relevant basis of analogy lies in the nature of the

power and not its object. The director of a corporation can only act as such through the

exercise of certain powers over the corporation. Directors’ powers, being residuary in

nature under most statutes of incorporation, are so broad that they are implicated in

virtually everything that directors do to effectuate the productive and other functions of

the corporation. They are implicated in the hiring of employees, the acquisition and

extension of credit, supplies, and property, the delegation of power to, and supervision of,

officers, decision-making on ordinary and extraordinary matters of business, and the

retention of, and issuance of instructions to, outside fiduciaries.

Directors have unusually broad powers. Nevertheless, they are fiduciary in the

sense described in Section 3.A. Directors’ powers are in the nature of authority derived

from the legal capacity of another person or group of persons. For corporations

incorporated under a general statute of incorporation, the powers are ordinarily bestowed

on directors by its express terms. These powers are, in turn, entailments of legislative

concession of legal personality to the corporation. Corporations receive legal personality

from the state but can only act on that personality – i.e. exercise particular legal

capacities – through directors and their delegates.

The powers of directors are derivative, discretionary, relational and specified. The

derivative nature of their powers follows from the recognition that no one enjoys the

power to represent an organization innately; they must be authorized to so act through

65
Adolf Berle, Corporate Powers as Powers in Trust, 44 HARV. L. REV. 1049 (1931); Merrick Dodd, For
Whom Are Corporate Managers Trustees? 45 HARV. L. REV. 1145 (1932); and Adolf Berle, For Whom
Corporate Managers Are Trustees, 45 HARV. L. REV. 1365 (1932).

32

Electronic copy available at: https://ssrn.com/abstract=2353062

legislative enactment or by the express will of its members. The derivative character of

directors’ powers is perhaps seen most clearly in their susceptibility to annulment or

reclamation. The state ordinarily reserves the right to dissolve corporations and/or to

remove and replace directors under general statutes of incorporation. Shareholders are

also ordinarily given the power to displace directors by voting them out in elections or by

asserting the authority to act in their stead under a shareholder agreement.

The specificity of directors’ powers may be less evident, given that they are

residual and granted in very broad terms.66 However, though broad, directors’ powers are

limited. The director may exercise only those powers that the corporation or its members

enjoy as such, for genuine corporate purposes, and only for the term of her office.67

Further, whether or not directors’ powers are actually specified, they are susceptible to

specification in constating documents either ex ante or through ex post amendment by

resolution of members of the corporation.

The discretionary nature of directors’ powers is well known. Directors are elected

by, and answerable to, shareholders. However, it is settled law that directors are not

bound to abide by the express will of shareholders on matters within their authority.68

Rather, they are free to exercise discretion in the best interests of the corporation. The

scope for discretion enjoyed by directors is in some jurisdictions underscored by the

business judgment rule. Formulations of the rule vary, but in general it provides that

66
Specificity is more evident in contractarian and special act corporations, in that directors of these
corporations have only those powers specifically conferred.
67
Berle, Corporate Powers, supra note 65; Re Smith and Fawcett Ltd [1942] Ch. 304; and Teck Corp. Ltd.
v. Millar (1972) 33 D.L.R. (3d) 288 (BCSC).
68
Automatic Self-Cleansing Filter Syndicate Co Ltd v. Cuninghame [1906] 2 Ch. 34 (C.A.); and Scott v.
Scott [1943] 1 All E.R. 582 (Ch.D.).

33

Electronic copy available at: https://ssrn.com/abstract=2353062

powers of judicial review may not be exercised so as to erode the discretion afforded to

directors in the exercise of their powers.

Directors’ powers are also relational. Consider first an old variant on the

corporate form - the contractarian (or memorandum) corporation, wherein directors act

for shareholders in aggregate on the basis of their authorization. Shareholders’

appointment of a director transforms the basis on which she interacts with them. The

director exercises power relationally in that she must be taken to be acting on behalf of

shareholders, jointly and severally, by virtue of their membership of the corporation.

Personified corporations complicate the analysis, but only slightly. Here, the fact that

directors’ powers are bestowed by the state over a statutorily created entity gives

directorship the appearance of a public office. However, appearances deceive. For all of

their similarities to public offices, corporate offices are private, and powers associated

with them are wielded relationally in respect of constituents of the organization itself.

Personification means only that directors derive their powers from, and wield them

directly in relation to, the organization rather than its members. These powers alter the

terms on which directors and the organization (including shareholders, in and through

identification of their interests with those of the personified entity) interact, for it is only

through fiduciary administration that the organization is able to act as person.

5. Testing the Fiduciary Powers Theory: Hard Cases

A basic test of the functionality of a definition is whether it permits identification of

exemplars that one believes ex ante should fall within an adequately stipulated definition.

Our core cases, above, provided such a test for the definition of the fiduciary relationship

34

Electronic copy available at: https://ssrn.com/abstract=2353062

stipulated by the fiduciary powers theory. But of course a basic test is not necessarily the

best test. In the real world, identification of particulars as being of a defined kind or

category is not always simple. A more demanding test is that of identification through

differentiation in hard cases (i.e., cases requiring one to sort out potentially confounding

similarities or differences).69

As it happens, there are several hard cases of relationship characterization in

fiduciary law. Courts and commentators have disagreed, sometimes vehemently, about

whether a particular sort of person, office, or relationship should be considered fiduciary.

The fiduciary powers theory promises principled resolution of these debates. I here

consider three hard cases. The first is whether advisors should, in general, be considered

fiduciaries by virtue of the properties of advisory relationships in general. The second is

whether confidants should, again in general, be considered fiduciaries. The third and last

involves public officials. Here, the question is whether there is something in the nature of

a public office or of state authority that would suggest that public officials are fiduciaries.

69
Cases that are hard in this sense have proven difficult to resolve because confounding similarities strain
working criteria of kind-status upon which differentiation is based. The cases have proven hard to resolve
through shared experience over time. Of course, cases could be hard in another sense; there might be
relationships that are fiduciary as a matter of custom and consensus but seem difficult to explain in light of
the fiduciary powers theory. Consider relationships in which the fiduciary wields power over the interests
of multiple beneficiaries. In such cases, the existence of a group of beneficiaries complicates the idea that
the fiduciary derives power from a process of authorization that involves transfer of capacity. The transfer
of capacity is readily identified in cases in which there is one fiduciary, one beneficiary (and, perhaps, one
benefactor). It is less readily identified otherwise. Consider as well the relationship between parents and
children. In many, if not most, common law jurisdictions, the parent-child relationship is considered
fiduciary as a matter of status. Parents wield discretionary power over their children. However, children
lack capacity. Thus, though the law might insist (or imply) that the parent be understood as having derived
power from the legal personality of the child, the parent’s power cannot be understood as taking the form of
a capacity of the child because the child’s personality does not entail (realized) capacities. I believe that the
fiduciary powers theory can account for these cases, however neither is a hard case in the sense used here.
They test the limits of the explanatory potential of the theory but they have no bearing on its capacity to
resolve controversies over the fiduciary character of certain relationships. I am grateful to Evan Fox-Decent
for raising this issue.

35

Electronic copy available at: https://ssrn.com/abstract=2353062

A. Advisors as Fiduciaries

It is sometimes thought that all manner of advisors – lawyers, doctors, bankers,

tax and investment advisors, and others – are fiduciaries of their clients simply by virtue

of their advisory function.70 This characterization seems supportable on the basis of some

loose comparisons. Like trustees, certain advisors have access to the property of others.

Many are expected to exercise discretion in advising their clients. Advisory relationships

are frequently characterized by trust, confidence, inequality, and vulnerability.71 Why,

then, should we not say that advisors in general are fiduciaries of their clients?

One might think we should not say this because it is not true.72 Loose

comparisons provide insufficient support for general characterization, particularly where

the characterization is so important. The question of the fiduciary character of advisory

relationships arose in Hodgkinson v. Simms, a case in which the Canadian Supreme Court

had to decide whether a financial advisor was a fiduciary in the provision of tax and

investment advice.73 The Court was divided on several issues but was unanimous in

holding that an advisory relationship is not inherently fiduciary. Justice La Forest, writing

for the majority, explained that for an advisory relationship to be fiduciary “there must be

something more than a simple undertaking by one party to provide information and execute

orders for the other.”74 The dissenting justices agreed. Justice McLachlin cautioned that: “a

false indicator of a fiduciary obligation is the “category” into which the relationship falls.

Professional relationships like doctor-patient and lawyer-client often possess fiduciary

70
J.C. SHEPHERD, THE LAW OF FIDUCIARIES 28 (1984): “any person can, by offering to give advice in a
particular manner to another, create in himself fiduciary obligations stemming from the confidential nature of the
relationship created, which obligations limit the adviser's dealings with the advisee.”
71
Kenneth M. Lodge and Thomas J. Cunningham, The Banker as Inadvertent Fiduciary: Beware a
Borrower’s Special Trust and Confidence, 98 COMM. L.J. 277 (1993).
72
Jill E. Fisch, Fiduciary Duties and the Analyst Scandals, 58 ALA. L. REV. 1083 (2007).
73
[1994] 3 SCR 377 (SCC).
74
Id. at 409-410.

36

Electronic copy available at: https://ssrn.com/abstract=2353062

aspects. But equally, many of the tasks undertaken pursuant to these relationships may not be

trust-like or attract a fiduciary obligation.”75

The fiduciary powers theory provides necessary conceptual support for this view.

It suggests that advisors are not fiduciaries as such. That is to say, advisors are not

fiduciaries by virtue of giving advice.76 Instead, they are fiduciaries only where they

exercise discretionary power over the practical interests of their clients. In such cases,

provision of advice is incidental to the exercise of discretionary power.

This is consistent with cases that have investigated the circumstances in which

advisory relationships are, or may become, fiduciary. For example, in Hodgkinson,

Justice McLachlin said that “the cases suggest that the distinguishing characteristic between

advice simpliciter and advice giving rise to a fiduciary duty is the ceding by one party of

effective power to the other. It is this mutual conferring and acceptance of power to the

knowledge of both parties that creates the special and onerous trust obligation.”77 To much

the same effect is a distinction drawn in U.S. law between investment advisors and

brokers. Investment advisors are assumed to be fiduciaries insofar as they are presumed

to enjoy discretionary power in making investment decisions for clients; brokers are not

assumed to be fiduciaries as it is presumed that they lack such power. Jill Fisch explains:

“as a general matter, brokers do not owe broad fiduciary obligations to their customers …

[but] special circumstances such as investor allocation of discretionary authority to the

75
Id. at 464.
76
Save for the rare circumstance in which the advisee is so epistemically dependent on the advisor that he
is incapable of exercising independent judgment in determining how to act in reliance on the advice. In
cases characterized by total reliance the advisor enjoys effective discretionary power over the advisee;
though there is no formal cession of capacity the advisee’s exercise of it is effectively determined by the
advice given.
77
Supra note 74, at 466 (emphasis supplied).

37

Electronic copy available at: https://ssrn.com/abstract=2353062

broker may lead to the imposition of fiduciary principles.”78 Similarly, Arthur Laby

observes that “[m]any courts adhere to a general rule that brokers are not subject to

fiduciary duties unless they have investment discretion over an account.”79

The fiduciary powers theory enables may assist in the resolution of debate over

the fiduciary character of particular kinds of advisory relationships. Fiduciary

characterization of doctor-patient relationships has, for example, proven to be highly

controversial.80 The fiduciary powers theory provides a basis upon which to differentiate

amongst doctor-patient relationships. Where services to be performed are such that the

doctor must seek authorization to act on the patient (e.g., to perform an invasive

procedure) and this will necessarily require her to exercise independent judgment, the

doctor is a fiduciary. However, where the doctor’s role is limited to providing

information to enable the patient to exercise independent judgment or where the doctor

provides information to a third party she exercises no power per se and is not a fiduciary.

B. Confidants as Fiduciaries

It is also commonly suggested that confidants are or should be considered

fiduciaries. Sometimes this is a matter of semantic slippage,81 “confidence” being

understood as synonymous with “trust” or “reliance.”82 To place confidence in another in

78
Supra note 72, at 1094-1095.
79
Arthur Laby, Fiduciary Obligations of Broker-Dealers, 55 VILL. L. REV. 701, 704-705 (2010). See also
Donald C. Langevoort, Brokers as Fiduciaries, 71 U. PITT. L. REV. 439 (2010).
80
Norberg v. Wynrib, supra note 32; Breen v. Williams, supra note 7; Peter Bartlett, Doctors as
Fiduciaries: Equitable Regulation of the Doctor-Patient Relationship, 5 MOD. L.R. 193 (1997).
81
See generally Dennis Klinck, Things of Confidence: Loyalty, Secrecy and Fiduciary Obligation, 54
SASK. L. REV. 73 (1990).
82
Id. at 77-79, citing by way of illustration, Billage v. Southee (1852) 9 Hare 534, 540 (68 E.R. 623) (Ch.),
in which Sir George Turner said: “No part of the jurisdiction of the Court is more useful than that which it
exercises in watching and controlling transactions between persons standing in a relation of confidence to
each other … The jurisdiction is founded on the principle of correcting abuses of confidence and I shall

38

Electronic copy available at: https://ssrn.com/abstract=2353062

this sense is to trust another with some interest one cares about and/or to rely upon them

to have a certain attitude to, or inclination to act in a certain way toward, you. For present

purposes, I am interested in the distinct suggestion that relationships of confidence are

fiduciary, where “confidence” consists in the disclosure and knowing receipt of

confidential information. A number of courts and commentators have suggested that

confidants in this sense are fiduciaries.83 And here, too, the characterization seems

supportable on the basis of loose comparisons. A confidant, like a trustee, has access to a

“resource” (information) that “belongs” to someone else. Furthermore, relationships of

confidence, like many fiduciary relationships, involve trust or engender dependence.

Nevertheless, others have insisted that confidants are not fiduciaries.84

The fiduciary powers theory may assist in the resolution of this controversy as

well. As was true of advisors, it suggests that confidants are not fiduciaries as such; i.e.,

by virtue of their knowing receipt of information disclosed in confidence. Rather,

confidants are fiduciaries wherever they happen to wield discretionary power relative to

the practical interests of a person who might also have made a disclosure of confidential

have no hesitation in saying that it ought to be applied, whatever may be the nature of the confidence
reposed or the relation of the parties between whom it has subsisted. I take the principle to be one of
universal application, and the cases in which the jurisdiction has been exercised – thus of trustee and cestui
que trust – guardian and ward – attorney and client – surgeon and patient – to be merely instances of the
application of the principle.”
83
See Lac Minerals, supra note 26. per Wilson J (according to whom a fiduciary relationship was
established between two companies negotiating toward a joint venture on the basis of disclosure of
confidential information relating to the venture by one party to the other); Finn, supra note 11, at 36
“[disclosure of] confidential information … will attract fiduciary law’s protection in Commonwealth
jurisdictions provided the circumstances are such as to give rise to a duty of confidence”; R.A. Brait, The
Unauthorized Use of Confidential Information, 18 CAN. BUS. L.J. 323, 336 (1991); Robert Flannigan, The
Fiduciary Obligation, 9 OXFORD J. LEGAL STUD. 285, 286 (1989); and Daniel Bayliss, Breach of
Confidence as a Breach of Fiduciary Obligations: A Theory, 9 AUCK. U. L. REV. 702 (2000).
84
See Lac Minerals, id., per Sopinka J. (dissenting) at 600: “the fact that confidential information is
obtained and misused cannot of itself create a fiduciary obligation”; R.G. Hammond, Is Breach of
Confidence Properly Analyzed in Fiduciary Terms? 25 MCGILL L.J. 244 (1979); and Tamar Frankel,
Fiduciary Law, 71 CAL. L. REV. 795, 825 (1983): explaining that it “is evident that while [fiduciary and
confidential] relationships may exist simultaneously, they do not necessarily do so.”. Trust and confidence
alone are not grounds for imposing fiduciary duties on the confidant,” citing Vai v. Bank of America, 56
Cal. 2d 329, 337-338 (1961).

39

Electronic copy available at: https://ssrn.com/abstract=2353062

information. Invariably, where the exercise of power entails disclosure and use of

confidential information, fiduciary regulation of power will constrain the handling of the

information. But these consequences are incidental.

It should be recognized that fiduciaries are often in receipt of confidential

information.85 Indeed, disclosure of such information is effectively necessary for proper

discharge of some fiduciary mandates. Nevertheless, as a conceptual matter, the

disclosure and receipt of confidential information does nothing to alter the basis for

considering a relationship to be fiduciary. One is a fiduciary by virtue of one’s possession

of fiduciary power, regardless whether the mandate under which one receives such

power, or circumstances in which one exercises it, also happen to involve disclosure of

confidential information.

C. Public Officials as Fiduciaries

Thus far, we have been considering the fiduciary relationship as a distinctive kind of

private law relationship. This is appropriate enough, for fiduciary duties are first and

foremost a manifestation of private right. They govern the administration of certain

private institutions and relationships in civil society. Nevertheless, one might wonder

whether, and if so under what circumstances, public officials might be fiduciaries.

There is a large and growing body of work in public law theory that argues that

public officials are fiduciaries.86 The argument has been developed in different ways. For

example, Evan Criddle claims that the administrative state is inherently fiduciary and that

85
See Klinck, supra note 81, at 80: “Fiduciaries often are given, or are placed in a position that gives them
access to, information.”
86
Though prominent private law scholars have also argued that public officials are fiduciaries. See Paul
Finn, The Forgotten “Trust”: The People and the State, in EQUITY: ISSUES AND TRENDS (M. Cope ed.,
Federation Press 1995) and FRANKEL, supra note 4.

40

Electronic copy available at: https://ssrn.com/abstract=2353062

principles of administrative law may be usefully reinterpreted in this light.87 Teddy Rave

argues that politicians should be considered fiduciaries and thus be prohibited from

gerrymandering.88 Sung Hui Kim argues that the fiduciary nature of political

representation entails limits on Congressional insider trading.89 Ethan Leib contends that

judges are fiduciaries and that this has important implications for our understanding of

principles of adjudication, judicial ethics, and the place of the judiciary in democratic

government.90 Evan Fox-Decent provides a richly developed argument that all manner of

public officials are fiduciaries in his book, Sovereignty’s Promise.91 Fox-Decent argues

that public officials are fiduciaries insofar as the execution of official functions rests on

an assertion of sovereign power that can only be understood as legitimate if conditioned

in fiduciary terms.92

Claims that public officials are fiduciaries have met with some skepticism. As

Fox-Decent explains, “[m]any common law courts have shown considerable reluctance to

characterize as fiduciary the relationship between front-line decision-makers and the

individuals subject to them.”93 For instance, in Harris v Canada, Dawson J held that a

federal Minister could not be a fiduciary lest he face conflicting and irreconcilable duties

to multiple beneficiaries.94 More recently, in Alberta v Elder Advocates of Alberta

87
Evan J. Criddle, Fiduciary Foundations of Administrative Law, 54 UCLA L. REV. 117 (2006).
88
D. Theodore Rave, Politicians as Fiduciaries, 126 HARV. L. REV. 671 (2013).
89
Sung H.. Kim, The Last Temptation of Congress: Legislator Insider Trading and the Fiduciary Norm
Against Corruption, 98 CORNELL L. REV. 845 (2013).
90
Ethan J. Leib, David L. Ponet, and Michael Serota, A Fiduciary Theory of Judging, 101 CAL. L. REV. 699
(2013).
91
FOX-DECENT, supra note 32.
92
Relying on Wilson J.’s analysis of fiduciary relationships in Frame v Smith, Fox-Decent argues more
specifically that public officials are fiduciaries to the extent that they: a) wield power; b) relative to
practical interests of other persons; c) in circumstances in which those other persons are vulnerable in the
sense of being incapable of exercising or controlling the power; d) and thus in circumstances characterized
by presumed trust. Id. at 89-112.
93
Id. at 152.
94
2001 F.C.R. 1408 (F.C.T.D.)

41

Electronic copy available at: https://ssrn.com/abstract=2353062

Society, McLachlin CJ stated that: “the special characteristics of governmental

responsibilities and functions mean that governments will owe fiduciary duties only in

limited and special circumstances”95 and further that “no fiduciary duty is owed to the

public as a whole.”96 The Chief Justice reiterated concerns voiced by Dawson J in Harris,

reasoning that imposing a duty of loyalty on the state would be “inherently at odds with

its duty to act in the best interests of society as a whole, and its obligation to spread

limited resources among competing groups with equally valid claims to its assistance.”97

The decisions in Harris and Elder Advocates reflect a perceived problem of

translation from private to public law; namely, how to reconcile public officials’ duty to

be responsive to the diverse (and often conflicting) interests of all citizens with a duty of

loyalty giving rise to a private right of action by individual citizens demanding special

consideration of their interests. This problem, though real, misses the mark. It is rarely

argued that the fiduciary constitution of public office is such that it generates private law

duties, breach of which may give rise to a private right of action. Public law theorists

instead argue that recognition that public officials are fiduciaries offers valuable

interpretive and normative perspectives on established principles of public law (e.g.,

constraints on administrative discretion) or positions of public policy (e.g., limiting

gerrymandering). To the extent that private law rights and duties are discussed at all, it is

to suggest that public law principles are animated by cognate concerns (e.g., the concern

over self-interested abuse of power).

95
[2011] 2 S.C.R. 261 (S.C.C.) at para. 37.
96
Id. at para. 50.
97
Id. at para. 44. The perceived inconsistency between public mandates and the duty of loyalty may be
attributable in part to an overly narrow view of judges about the content of the duty of loyalty. See Andrew
Gold, The Loyalties of Fiduciary Law, in this volume. See also Andrew Gold, Reflections on the State as
Fiduciary, 63 U. TORONTO L.J. 655 (2013).

42

Electronic copy available at: https://ssrn.com/abstract=2353062

 As was true of our hard cases in private law, uncertainty over the appropriateness

of characterizing public officials as fiduciaries is hard to resolve without clarity on the

nature of the fiduciary relationship itself.98 Under the fiduciary powers theory, it is

important to distinguish two ways in which public officials may be thought to be

fiduciaries. First, a public official might be said to be a fiduciary of an individual or

defined group in just the way that ordinary private fiduciaries are, with the implication

that private law duties enforceable by private right of action apply to the official. Second,

one might say that a public official is a sui generis fiduciary of the public given the

nature of the relationship between the state and its people. This kind of claim does not

necessarily imply that public officials are subject to private law duties. Rather, it may

instead be that the fiduciary character of public office informs existing principles of

public law (enforceable, if at all, in the usual ways). An advantage of the fiduciary

powers theory is that it permits us to make this distinction and to evaluate each claim

about the fiduciary character of public office in its own right.

The court in Elder Advocates was surely right to deny that the state is in general a

fiduciary of the public in the ordinary sense; i.e., in a relationship giving rise to private

law duties enforceable by way of a private right of action. That does not mean that public

officials are not sui generis fiduciaries of the public. But for the moment, let us consider

when public officials might be fiduciaries in the ordinary sense. Elder Advocates clearly

98
Many who claim that public officials are fiduciaries say very little about fiduciary relationships in
general. However, Fox-Decent offers a particularly detailed account of the nature of the fiduciary
relationship. We agree that the fiduciary relationship involves exercise of discretionary power by one
person relative to another. However, there are important differences in our accounts. We both understand
fiduciary power as a form of authority, but Fox-Decent has not argued that fiduciary power is rooted in the
concept of legal personality. At times, he appears to suggest that fiduciary authority may be in the form of
factual or legal power. I also believe that some of the characteristics Fox-Decent ascribes to fiduciary
power are inessential (e.g., its allegedly administrative and institutional nature). Furthermore, Fox-Decent
suggest that trust and vulnerability are amongst the definitive properties of the fiduciary relationship, and
that they are pertinent to the legal justification for fiduciary duties, both propositions that I resist.

43

Electronic copy available at: https://ssrn.com/abstract=2353062

contemplates that there will be such cases. McLachlin CJ said that: “generally speaking, a

strong correspondence with one of the traditional categories of fiduciary relationship …

is a precondition to finding an implied fiduciary duty on the government.”99 She also

explained that an aggrieved beneficiary must establish “a specific private law interest to

which the person has a pre-existing distinct and complete legal entitlement.”100 Finally,

and most intriguingly, she held that the power wielded by the state relative to the

beneficiary must of a direct (nature:

[T]he degree of control exerted by the government over the interest in question
must be equivalent or analogous to direct administration of that interest before a
fiduciary relationship can be said to arise. The type of legal control over an
interest that arises from the ordinary exercise of statutory powers does not suffice.
Otherwise, fiduciary obligations would arise in most day to day government
functions making general action for the public good difficult or almost
impossible.101

These conditions are consistent with the fiduciary powers theory of the fiduciary

relationship. A public official may be a fiduciary of an individual or defined group only

where it enjoys discretionary power directly in relation to the practical interests of that

individual or group. In these cases, the official must have undertaken a mandate under

which it holds powers identical in kind to those held by private fiduciaries. These powers,

again, are in the nature of legal capacities ordinarily derived from the legal personality of

a particular individual beneficiary or benefactor (or group of same). Fiduciary powers in

this sense are not capacities associated with the sui generis legal personality of the state

(i.e., its legislative, administrative, police, or judicial powers). Rather, they are powers of

legal personality simpliciter (e.g., capacities to make, vary or enforce contracts, to

manage, alienate or invest property, and so on). Ordinarily, these powers are conferred on

99
Supra note 95, at para. 47.
100
Id. at para. 51.
101
Id. at para. 53.

44

Electronic copy available at: https://ssrn.com/abstract=2353062

private persons acting as fiduciaries. However, in certain instances, the state might

assume them. For instance, the state might undertake powers on trust to administer lands

for aboriginal bands;102 it might act as executor or trustee of last resort, administering

property of those who die intestate or fail to make a valid appointment of an executor or

trustee; it might undertake powers to invest pension funds for veterans;103 or it might

undertake powers of guardianship over children abandoned by or removed from their

families.104

This suggests that there is nothing distinctively public about certain mandates

undertaken by public officials. A public official just might undertake powers relative to

specific individuals or groups that have an inherently fiduciary character. The state in

these cases acts as trustee, agent, executor, or guardian. Where it does so, it may exercise

certain prerogative powers of the state (e.g., in assert jurisdiction over neglected or

abused children), but the fiduciary powers are indistinguishable from those wielded by

private fiduciaries. Being held for, and wielded relative to, a defined individual or group,

the state owes ordinary private law duties to act in the interests of the individual or group

in the exercise of its fiduciary powers.

Let us now turn to the claim that public offices are inherently fiduciary. We have

established that fiduciary power is a form of authority in the form of legal capacities

derived from the legal personality of an ordinary person. As just noted, the state can and

does from time to time undertake power in this sense. Might we say that public officials

in general are fiduciaries to the extent that exercise of official functions invariably entails

102
Guerin v. The Queen, supra note 25.
103
Authorson v. Canada (Attorney General) (2002) 58 O.R. (3d) 417; rev’d (on other grounds) 2003 S.C.C.
39.
104
K.L.B. v. British Columbia [2003] 2 S.C.R. 403 (S.C.C.).

45

Electronic copy available at: https://ssrn.com/abstract=2353062

exercise of capacities inherent in the legal personality of the state? While this line of

argument seems promising, it presents one obvious problem. The personality of the state

is sui generis. The personality of the state parallels that of private persons to the extent

that (a) it consists of certain legal capacities and (b) some of these capacities are the very

capacities that individuals have. However, the parallels are limited. The state is, in

addition to powers of ordinary personality, possessed of powers of a fundamentally

different character. These are powers of statehood that permit the state to function as

such; i.e., as an organization capable of making, giving and enforcing law. The basic

categories of powers of statehood bespeak the distinctiveness of the personality of the

state. They include legislative powers (i.e., the power to make laws of general

application), administrative powers (e.g., the power to administer the state through

bureaucracy and to administer revenue-generating and benefit-conferring programs),

enforcement powers (i.e., the power to ensure compliance with laws through policing and

prosecution), and judicial powers (e.g., the power to adjudicate civil, criminal, and other

legal matters). If public offices are inherently fiduciary, it cannot be because powers of

statehood are identical to powers of ordinary personality. This being so, it is necessary to

consider the extent to which powers of statehood and ordinary personality share common

properties.

Proper exploration of these commonalities will be intricate; it calls for more

nuanced analysis of the concepts of ordinary and state personality than I am able to

provide here. Nevertheless, a plausible line of argument might emphasize that both kinds

of power are ultimately rooted in parallel forms of legal personality that reflect

individuals’ capacity for self-determination. Powers of ordinary personality are the means

46

Electronic copy available at: https://ssrn.com/abstract=2353062

by which individuals are self-determining in law; i.e., by which they may fix and pursue

their own ends and are recognized as responsible as such. Powers of statehood relate to

the capacity of individuals to demand establishment of a rightful condition through the

person of the state. To the extent that the personality and authority of the state must be

understood as derived ultimately from our moral capacity for self-determination, powers

of statehood may well be of a kind with powers of ordinary personality. Both forms of

power are essentially representative in character. They are held on the footing that they

are to be exercised for and on behalf of another and in their interests alone.

6. Conclusion

Fiduciary liability is premised on the nature of the relationship between the

fiduciary and beneficiary. However, leading scholars have suggested that it is

meaningless to talk of fiduciary relationships as such; the fiduciary relationship is not a

distinctive and coherent general kind of legal relationship. An implication is that

fiduciary principles can, and must, be understood in abstraction from analysis of general

characteristics of the relationships in which they arise.

I have here argued that this view is wrong. The fiduciary relationship is a

distinctive and coherent general kind of legal relationship. It is definable. Furthermore, it

is important that it be defined, given the central role played by the construct in the

conceptual structure of fiduciary law.

Beyond arguing that the fiduciary relationship is definable, and that it should be

defined, I have here offered my own definition and associated theory of the fiduciary

relationship. The fiduciary powers theory suggests that the essential characteristic of all

47

Electronic copy available at: https://ssrn.com/abstract=2353062

fiduciary relationships is possession by the fiduciary of discretionary power over the

practical interests of the beneficiary. Fiduciary power takes the form of legal capacities

derived from the legal personality of persons (ordinarily, the personality of a person other

than the fiduciary - a beneficiary, or a third-party benefactor). I have shown that the

fiduciary powers theory can explain the presumptive fiduciary character of relationships

of recognized fiduciary status, and that it promises to advance debates over the fiduciary

character of other relationships. However perhaps the most important demonstration of

the value of the fiduciary powers theory – its capacity to support a unified theory of

fiduciary law – must await further analysis of the significance of fiduciary relationships

in fiduciary law.

48

Electronic copy available at: https://ssrn.com/abstract=2353062