Investor-consumer or overall welfare: Searching for the paradigm of recent reforms in financial services contracts (EUI Working Paper, with Prof. Grundmann)

LAW 2017/05 Department of Law Investor-consumer or overall welfare: Searching for the paradigm of recent reforms in financial services contracts Stefan Grundmann and Fabrizio Esposito Electronic copy available at: https://ssrn.com/abstract=2925794 Electronic copy available at: https://ssrn.com/abstract=2925794 European University Institute Department of Law INVESTOR-CONSUMER OR OVERALL WELFARE: SEARCHING FOR THE PARADIGM OF RECENT REFORMS IN FINANCIAL SERVICES CONTRACTS Stefan Grundmann and Fabrizio Esposito EUI Working Paper LAW 2017/05 Electronic copy available at: https://ssrn.com/abstract=2925794 This text may be downloaded for personal research purposes only. Any additional reproduction for other purposes, whether in hard copy or electronically, requires the consent of the authors. If cited or quoted, reference should be made to the full name of the authors, the title, the working paper or other series, the year, and the publisher. ISSN 1725-6739 © Stefan Grundmann and Fabrizio Esposito, 2017 Printed in Italy European University Institute Badia Fiesolana I-50014 San Domenico di Fiesole (FI) Italy www.eui.eu cadmus.eui.eu Abstract This article challenges the assumption that from an economic perspective the distribution of value (be it surplus or well-being) between parties in voluntary transactions matters at most for instrumental reasons—that is, for increasing the size of the pie. This assumption is probably the stronghold of institutional economics and traditional law and economics which most blatantly deviates from any legal perspective. A strain of economic literature going back at least to Adam Smith conceives of market relations as principal-agents relations between consumers and producers. As consumer welfare is not the only societal value, conflicts with other values can justify its sacrifice in some occasions. A just society cannot be reduced to an allocatively optimal general equilibrium. This article focuses on the development of two arguments in support of this framework. First, a coherence (or conceptual) argument. There is an astonishingly unexplored connection between the role of the consumer on the market and the investor in the firm—not to be confused with the “producers are consumers too” argument. Second, we build a principal multi agent framework where aggregate consumer welfare is the principal and consumers, entrepreneurs and the legal system are the agent. With this framework, we offer an explanation of central elements of the recently reformed EU financial services system, in particular fiduciary duties and the duty of responsible lending. Keywords Consumer sovereignty, principal-agent, market rationale, financial services, consumer credit. JEL Classification D61, D63, D86, G38, K10, K12. Author contact details: Stefan Grundmann Professor of Transnational Private Law European University Institute Florence, Italy Stefan.Grundmann@eui.eu Fabrizio Esposito PhD Researcher European University Institute Florence, Italy Fabrizio.Esposito@eui.eu Table of contents INTRODUCTION ............................................................................................................................ 1 INVESTOR-CONSUMER OR OVERALL WELFARE, THIS IS THE QUESTION ........................................ 4 The investment contract – pivotal for financial law reform, the definition of principal goals and the idea of governance .................................................................................................................. 4 Consumers and investors as principals… and beyond ........................................................................ 5 REGULATED MARKET BEHAVIOUR AND ALLOCATIVE EFFICIENCY: A PRINCIPAL MULTI AGENT FRAMEWORK ....................................................................................................... 12 Justifying the delegation model in principal-agent theory ................................................................ 12 Building the principal multi agent framework .................................................................................. 16 The principal-agent structure of EU consumer and corporate governance policies .......................... 19 MIFID II AND THE CLIENT RELATIONSHIP AS FIDUCIARY RELATIONSHIP: THE NEW DISCLOSURE AND KNOW YOUR CUSTOMER REGIME .................................................... 22 The models of explanation of fiduciary relationship ......................................................................... 22 Disclosure regime, know your customer and best execution rules as significant examples ............. 25 MIFID II AND ITS NEW CONFLICTS OF INTEREST REGIME........................................................... 27 The relationship to the basic fiduciary model ................................................................................... 27 First example: far-reaching organizational duties ............................................................................. 27 Second example: The duty to disclose remaining conflicts of interest ............................................. 28 Third example: The regime on fees (inducements) ........................................................................... 29 Conflicts of interest regulation within the principal multi agent framework .................................... 30 UNDERSTANDING THE REFORMED EU CONSUMER CREDIT LAW ................................................ 30 The aims of credit regulation in the EU ............................................................................................ 31 Worthy for whom? Three notions of responsible lending ................................................................. 32 EU regulated credit market as a principal multi agent interaction .................................................... 34 CONCLUSIONS ........................................................................................................................... 36 Introduction* The world has lived the deepest financial crisis since 1929, and among the rather numerous really important causes, two were prominent which clearly had to do with contract law: subprime lending (to consumers) formed the raw material from which the toxic papers were conceived (‘structured’);1 and lack of transparency of those papers (exacerbated by misleading AAA ratings) was seen as the main source of over-optimistic and poorly diversified investment strategies of market participants of all kinds. It is therefore not astonishing that also contract law – and very prominently EU contract law – was subjected to fundamental change in these fields: in the area of consumer credits with the thorough reform of the EU Consumer Credit Directive and the (completely new) adoption of a (Consumer) Mortgage Credit Directive introducing, among others, for the first time also a duty of responsible lending,2 and in the area of investment services with the fundamental reform of the Markets in Financial Instruments Directive (MiFID, now MiFID II), introducing many new instruments, among them also product related governance rules.3 These reforms are so rich, contain so many single new solutions and are so disputed that analysing them from the perspective of their efficiency enhancing potential would already provide enough material not for a lengthy paper, but for a comprehensive monograph.4 Instead, we propose to see these reform measures and the most striking new solutions found as paradigmatic and exemplary cases. We think that, again, banking law, namely banking contract law, can be seen as a motor of private law more generally,5 namely also of consumer contract law. If financial services contract law is therefore the subject matter of this article, there is a third reason for choosing this area – besides fundamental reform (leading to what at least the EU legislature understood as one the examples of highly cutting edge legislation) and besides the paradigmatic and innovative power of financial services law for all private and contract law: There are probably not many areas of * An early draft was presented at the EALE Conference 2016 in Bologna. We are especially thankful to our discussant, Prof. Fernando Gómez, for his inspiring and thoughtful comments. A refined version was presented at the SIDE Conference 2016 in Turin and later benefited for comments by Giorgio Monti. The usual disclaimer applies. 1 See, for instance, Mian and Sufi (2014); and on combatting subprime lending: see Y. Atamer (2011: 179-202). 2 See (reformed) Directive 2008/48/EC of the European Parliament and of the Council of 23 April 2008 on credit agreements for consumers and repealing Council Directive 87/102/EEC, OJEC 2008 L 133/66; this directive (as the one of 1987 – exempted mortgage backed loans which only have been covered by: Directive 2014/17/EU of the European Parliament and of the Council of 4 February 2014 on credit agreements for consumers relating to residential immovable property and amending Directives 2008/48/EC and 2013/36/EU and Regulation (EU) No 1093/2010, OJEC 2014 L 60/34, and only in the latter directive, a clear rule on the duty of responsible lending can be found – in Art. 18 para. 5 lit. a; see also Art. 7; for the duty of responsible lending see below, pp. 34-36. 3 Directive 2004/39/EC of the European Parliament and of the Council of 21 April 2004 on markets in financial instruments amending Council Directives 85/611/EEC and 93/6/EEC and Directive 2000/12/EC of the European Parliament and of the Council and repealing Council Directive 93/22/EEC [MiFID], OJEC 2004 L 145/1; now Directive 2014/65/EU of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Directive 2002/92/EC and Directive 2011/61/EU [MiFID II], OJEU 2014 L 173/349; and Regulation (EU) No 600/2014 of the European Parliament and of the Council of 15 May 2014 on markets in financial instruments and amending Regulation (EU) No 648/2012, OJEU 2014 L 173/84; for product governance see below, pp. 28-29. 4 For the MiFID II reform, see, for instance, Busch and Ferrarini (2016), and also below, pp. 24-32. 5 This is an idea dear namely to German doctrinal thinking (but not limited to it) where many examples could be given, but perhaps two very prominent ones should be sufficient: Claus-Wilhelm Canaris has written in the most prestigious large commentary – the so-called Staub’scher Großkommentar – his treatise on banking contract law (Canaris (1988)) and in this treatise, as the very first to do so, he describes the relevant types of contracts truly from ‘naissance to death’, from the precontractual phase to the treatment in insolvency, thus for the first time giving a full picture of all law related to contracts, here banking contracts. A second instance, where banking contract law opened a completely new chapter for contract law, is network of contracts, first described for payment systems: see Möschel (1986: 187). For today’s picture (where banking law is still the motor of private law in many other respects), see (in the new edition of the Staub’scher Großkommentar): Grundmann (2015, 2016, 2017, 5th edn.). 1 Stefan Grundmann and Fabrizio Esposito EU private and contract law where the images of different kinds of clients or consumers and the types of instruments for protection used are as manifold and as varied as in financial services law.6 Therefore, the legal material is not only cutting edge legislation, sensed as being particularly paradigmatic, but it is also extremely varied; a good basis for institutional comparison and normative assessment. This material will be used and analysed in two ways mainly. In a first step, we propose to revisit a foundational question: we will ask the question anew whether indeed the goal of markets and, more specifically, of market regulation should be seen in maximization of overall welfare – as would seem to be the mainstream view (today virtually undisputed) in economic theory.7 We will question this view and suggest that it is more convincing to identify two interrelated economic normative foundations of markets and of market regulation, namely a microeconomic and a macroeconomic foundation. At the macro level, the goal is growth (“the size of the pie”) – and we leave aside the thorny issue of its measurement. At the micro level, the goal is identified, consistently with the tradition, with allocative efficiency – but the foundational difference is that allocative efficiency is neither about wealth maximization nor overall or societal welfare maximization. In our perspective, a market is allocatively optimal when the overall welfare of the consumers of that market is maximized via mutual Pareto optimal transactions. Allocative efficiency is then a constrained optimization exercise where the constraints are consumers’ budget and a mutual Pareto optimality condition; a market allocation M a’ is thus allocatively more efficient than market allocation Ma iff the aggregate consumer welfare in Ma’ is superior to the aggregate consumer welfare in Ma and the Paretian constraint is respected. Notably, we do not claim that this two-pillar foundation is exhaustive of societal goals (see below, pp. 13-14). From a different perspective, it is pivotal to our understanding of allocative efficiency to conceive of the consumer/investor-entrepreneur/manager relation in terms of a principal-agent relation (see below, pp. 21-24). Accepting this theoretical reconstruction of the market rationale would quite fundamentally modify and change the analytical framework of any legal inquiry using economic theory.8 It would not only fundamentally change the analytical framework in the economic analysis, it would bring the analytical framework of economic theory also much closer to the dominant paradigm in actual EU legislation. Here, the dominant view would indeed be that the regulation of private autonomy is primarily justified by a concern for the unfair consequences of flatly enforcing contracts between formally equal, but substantially unequal parties.9 Thus, unfair terms are unenforceable because they are one-sided, and not because them being one-sided implies that they reduce overall welfare.10 An immediate consequence of this understanding of allocative efficiency is offering a standpoint for clarifying the interplay between the Single Market and consumer policy. Contrary to a view often held by EU lawyers, the consumer interest is not cunningly instrumentalized (or even distorted) by EU institutions in order to foster the Single Market. A market working for consumers is actually, together with growth, the main justification of this project. For example, in 2009, when the European Commission President, José Manuel Barroso, asked the former Commissioner Mario Monti to suggest a strategy for “relaunching the Single Market”, he identified the benefits of the Single Market in “a powerful engine for growth and delivering the full benefits to consumers”. From this perspective, the framework discussed in this paper actually allows to take the face value of declarations like this seriously. We think that the analysis of the aforementioned directives (MiFID II, CCD, and MCD) is particularly suited for serving as a concrete example for our broader claim. Applied to these market interactions, the microeconomic normative goal is the maximization of the welfare of investors and of borrowers under 6 On this topic, see the comprehensive collection of essays in Leczykiewicz and Weatherill (2016). 7 For a survey, see Esposito (forthcoming: chapter 2). 8 We consider the present research an example of Law and Economics, as recently defined by Guido Calabresi and contrasted by the author to Economic Analysis of Law. See below, pp. 38-40. 9 For example, Reich (2009: 6-8). 10 2 For a more detailed discussion of the disagreement, see below, pp. 7-8 and, more extensively, Esposito (2016). Department of Law Working Papers Investor-Consumer or Overall Welfare the Paretian constraint.11 The economic grounds of this first step cannot be fully developed in this article. Such conceptual premise is reserved to later, more extensive work.12 What can, however, be achieved here and what is important as a basis for the second step is to expose one of the core arguments which speaks in favour of such an alternative approach. This preliminary inquiry of the micro level ultimate goal is done below (pp. 7-14). In a second step, we propose to analyse centre-pieces of said reforms in financial services regulation at the EU level. To do so, we sketch a principal multi agent framework at pp. 14-21. The award of the 2016 “Nobel” Prize in Economics to Hart and Holmström makes this approach very salient from an economic perspective. However, we identify and discuss two important ambiguities in the principal-agent literature. First, sometimes the framework adopts the principal’s view, while in others it takes the social planner’s view—the difference being that the maximand is either the principal’s or societal welfare. Second, sometimes an interaction is framed as principal-agent according to the delegation criterion while in others according to the information criterion—the difference being that the principal is either the delegator or the party to whom an information or action is hidden. We argue in favour of a delegating principal’s welfare conception of principal-agent theory. In the framework we propose, the principal is aggregate consumer welfare and the agents are the consumers and the entrepreneurs (together: primary agents) and the legal system (acting as monitoring agent). In the application of this framework, we start from MiFID II – devoting two sections to it (pp. 24-32) – and do so for two reasons. First, as the investment advice activity is seen as subjected to fiduciary duties already in mainstream legal analysis,13 it would seem less ‘astonishing’ to analyse this reform in the terms proposed, i.e. as a principal agent relationship between client and services provider. It should, however, be kept in mind that investment advice transactions are typically accompanied also by execution, i.e. sales of securities, and therefore could actually be seen as ‘sales’ transactions. The framework we propose, as we shall see (pp. 24-29), allows to explain why in these sales the service provider has typical obligations of an agent, as well and that it is interesting to explain why they are so completely dominated by the fiduciary part of giving disclosure and good advice and even find the best deal when executing (‘best execution’). The second reason for focusing on MiFID II predominantly is the richness of the regime: combining organizational and transactional features, combining standards and rules with preventive measures, combining market considerations with considerations on the individual contract relationship and being in massive transformation from MiFID I to MiFID II in many of the instruments used. Therefore, even two sections on two core (sets of) instruments is by no means ‘excessive’. Perhaps even more important for our argument is the analysis of the recent lending reform in consumer credit law. In fact, while investment advice is seen to be conditioned by fiduciary duties already by mainstream legal analysis, the loan business is different in this respect: indeed lending is not seen as a fiduciary relationship – or principal agent relationship – in mainstream legal analysis.14 There may be a certain overspill from the large part of the banking contract business which is indeed seen as being fiduciary in kind,15 but loans are primarily exchange contracts. They therefore would perhaps be even more meaningful for the change of paradigm from overall welfare to consumer/client welfare – if really regulation there as well can be understood as aiming at consumer/client welfare primarily and as being well conceived if it does so. Therefore at least a few centre-pieces of the consumer credit law reforms in the EU serve as background and indeed as contrasting forms of regulation to the MiFID II reforms. This area is taken up at pp. 32-38. We conclude with an elaboration of the future steps of the research agenda sketched in this article (pp. 38-40). 11 See, below, pp. 14-32. 12 Esposito (forthcoming: chapters 1-3); an earlier version of the argument can be found in Esposito (2013: 160-162). 13 See, for instance, Gevurtz (2010, 2nd edn.: 340–407); Grundmann (2012: § 23); Shepherd (1981: 132–136); Scoles, Halbach and Roberts (2006, 7th edn.: 747–774, esp. 747). In English law, the more currently used term is that of „fairness“: see Blair and Walker (2006: para. 11.51). 14 See below, pp. 29-32. 15 For this consideration, see more extensively (with examples): Grundmann (2016, 5th edn.: Part 1 para. 6 and part 2 para. 1719, 72 et seq). European University Institute 3 Stefan Grundmann and Fabrizio Esposito Investor-consumer or overall welfare, this is the question The investment contract – pivotal for financial law reform, the definition of principal goals and the idea of governance The investment contract happens to be the subject of what is probably the most fundamental reform in financial services contract law in the aftermath of the global financial crisis – the transition from MiFID I to MiFID II. Therefore, the considerations in the more concrete sections below (pp. 24-32) will focus in good part on that regime and the reform steps taken there. The investment contract is, however, also highly interesting from the point of view of the theoretical foundations of this article, as it is the contract which sits right at the cross-road between choice on markets and choice within firms. The former is typically related to consumers or clients as decision makers, the latter to shareholders as decision makers – always, of course, in an interplay with their counterparts who typically are the same ones: entrepreneurs16 who act as suppliers on markets, albeit via delegation to employees in the firm, and as managers of the firm, again often also via delegation. One very foundational question – taken up below (pp. 7-14) – would seem to follow from this: is it convincing at all that shareholders are conceived as principals in the mainstream economic theory on decision making in firms, while consumers are not, at least would seem to be not in mainstream economic theory on decision making on markets? Before we approach this question – whether such a difference in conception of the role of directors when they transact with shareholders/investors and with clients/consumers does not lack consistence – one last point about the investment contract needs to be recalled: this contract is as well at the heart of what can be seen as the key architectural feature of corporate governance debate, i.e. that debate which is the paradigm of a genuine and intimate collaboration of law and economics in the centre of business law. It is the investment contract – also in the form of mass investment in takeovers, changing the control structure in the target company – which forms the strongest link between so-called internal and external corporate governance. Internal governance is about positive influence on managers’ decision making via instruments of internal organisation (such as voting power, liability, checks and balances between different organs etc.), while external governance is about the positive influence from outside, namely via market transactions, and the combination of both is to be optimized.17 Between internal and external governance, the investment contract and – when the transactions are so numerous that they trigger a change of control – the takeover form probably the strongest link. In fact, both the investment contract and the takeover reside in transactions of shares on markets, but deeply influence decision making within the company: because they influence stock prices which then motivate directors in manifold ways (bonuses, prestige, potentially positions on boards etc.) or because, even more directly, they lead to a change of managers, i.e. of the primary decision makers themselves.18 H. Manne who first described this link for takeovers and thereby ‘discovered’ the ‘market for corporate control’ succinctly summarizes the main thrust or benefit of these ‘market’ transactions by saying: “But the greatest benefits of the takeover scheme [those market transactions] probably inure to those least conscious of it [the 16 Typically, the entrepreneur in financial services is a corporation, embodied by its directors according to company law. On decision making procedures in European company law, see: Davies et al. (2013); Grundmann (with the partial contribution of Glasow) (2012: 247-275); Dorresteijn et al. (2009:157-202). 17 It is one of the core understandings of general corporate governance debate that internal and external corporate governance are not mutually excluding each other or independent from each other, but that their combination and the optimization of such combination are core aims of good corporate governance. See, for instance, Grundmann (2012, 2nd edn.: § 14); and (with good comparative law indications): Hopt (2000: 106 et seq.); Parkinson (1995: 178–199); Teichmann (2001: 646 et seq.); segregating those states which stress external mechanisms (apart from the Anglo-American world also France in tendency) from those which definitely put more weight on internal mechanisms (namely the other EU Member States, esp. Germany): Wymeersch (1995: 309–315); more recent comparative law survey also by Wymeersch (2006: 619-631). 18 Path breaking for this role of the investment contract and of takeovers (as mass dis-investment and re-investment), Diamond and Verrecchia (1982); Holmström and Tirole (1990), and, of course, Manne (next footnote); today Allen (1993: especially 87 et seqq.); Avgouleas, (2012: 24 et seqq). 4 Department of Law Working Papers Investor-Consumer or Overall Welfare shareholders exposed to management decisions]. Apart from the stock market, we have no objective standard of managerial efficiency”.19 Thus, while being based on instruments and while being executed outside the company’s organisation, these instruments change the internal composition of decision makers, the duties within the company (for instance the remuneration) or at least strongly influence decision making in the company. This strong link between market (transactions/exchanges) and firm (organisation) is as well the aspect which shall now be taken up with a view to re-visit the foundational question: Is it convincing and consistent at all that shareholders are conceived as principals in mainstream economic theory on decision making in firms, while consumers are not, at least would seem to be not in mainstream economic theory on decision making on markets? Consumers and investors as principals… and beyond At this point, four considerations need to be made to complete the first step of our discourse: a) that indeed consumers/clients are not seen as principals by mainstream economics while shareholders/investors are, b) that this is formally questionable as both roles are strongly linked and in fact can be seen as the two sides of the same coin, c) that this is also questionable from the substantive point of view of how principal agency theory justifies the position of shareholders as principals (with reasons which exactly apply also to the client/supplier relationship on markets), and d) finally, some consideration has to be given to the question how far consumer welfare goes, which are the limitations of this normative goal. Together, points a), b) and c) build an argumentative scheme composed of a coherency part and of an authoritative economic argument part. First, we call for a coherent account of the normative foundation of clients/consumers and entrepreneurs relation with the shareholders/investors and managers relation. Second, we point out that there is an economic authority argument suggesting that a coherent account can be given by taking the client/consumers and shareholders/investors welfare as the maximand (under the Paretian constraint). Point d) places these considerations in a broader normative context. We take up each of these considerations in turn: a) Indeed, the traditional view in mainstream economics would seem to be, on the one hand, that markets and regulation of markets serve – as their ultimate goal – to maximize overall welfare, not to maximize the welfare of one side of markets, namely that of clients/consumers: the market process is basically conceived of as a directionless maximizer of overall welfare.20 Directionless, because the distribution of welfare between the parties is of no relevance at all from a normative perspective. The argument is that distributive concerns are better addressed outside the market system.21 Thus, even “Consumer protection is not an end in itself, especially in competitive markets; it can be justified, if at all, solely as a means to maximize the net value for both parties”.22 On the other hand, it seems equally solid a mainstream position in economic theory today that managers of firms are to be seen as agents and shareholders as 19 Manne (1965: 113). 20 Mainstream market analysis adopts a specific unit of measure of welfare – surplus - introduced by Alfred Marshall and defined as “(t)he excess of the price which [the consumer] would be willing to pay rather than go without the thing, over that which he actually does pay, is the economic measure of this surplus satisfaction. It may be called consumer’s surplus” (Marshall 2013 (1920): 103). The surplus maximization claim is expressed in canonical formula as follows: “In comparison with the competitive outcome, monopoly involves a transfer from consumers to suppliers. But the efficiency loss consists solely of the reduction in Consumer and Producer Surplus that are due to the lessened volume of trade” Hirshleifer (1976: 287). Besides its insensitivity to distribution of welfare between the parties of the exchange, this analysis suffers of two main limitations. First, surplus is an indirect measure of the intensity of preference. Thus, if the consumption choice is uninformed or not fully rational, surplus is an unreliable measure of welfare. Second, surplus measurement suffers from the wealth effect: the preferences of the rich count more than the ones of the poor. 21 Seminal to this regard is Kaplow and Shavell (1994); see also Kaplow (2008) and the critical discussions in Jolls (1998), Adler (2012: 560-567) and Calabresi (2016: especially 74-77). A major problem, that cannot be discussed here, is that while Kaplow and Shavell’s argument is only about the redistribution of income, it is often taken to relate to the distribution of wealth more generally. 22 Epstein (2013: 219). European University Institute 5 Stefan Grundmann and Fabrizio Esposito principals. It may be that Berle/Means, when they first described the relationship from a law and economics perspective in 1932, could conceive of the relationship rather differently and still could famously write: “Our idea does not accord either with the popular or the legal concept of a shareholder [i.e. as an owner]. … The various incidental rights – voting, pre-emptive rights in new stock issues, and the like [the traditional bundle of shareholders’ ‘property’ rights] … [are] merely uncertain expectations in the hands of the individual”.23 No property position, just expectations. Here it is interesting to note a first normative symmetry between the shareholder and the consumer. They are both residual claimants. The shareholder is entitled to all the monetary benefits remaining after the other factors of production have been paid. Similarly, the consumer is entitled to all the welfare that consumption gives him once the market price of the good or service has been paid. In both cases, this residue can be high or low, or even negative. Indeed, principal agent theory gives a strange perspective or image of the real power relationship between managers and shareholders. It is unrealistic to conceive of a powerful CEO as a mere agent of a (group of dispersed) small investor(s), and a sociological perception could never be that of the shareholder acting in a role of ‘principal’. All this conceded, it can nevertheless not be neglected that the starting point taken in mainstream economics is still this: the whole of the theory of corporate finance and decision making, thereby also of corporate governance is based on a model in which the shareholder is the principal, endowed with ultimate decision making power, carrying the ultimate risk and having a right to the residual gain. And, as will be seen, there are very strong reasons for such a concept, especially if one considers the way incentives for the individual decision makers are framed by this concept. Therefore, the firm as an organization and its regulation is normatively aimed at – as ultimate goal – the furthering of the interest of the principal, that is, the investors.24 And again, there is a normative symmetry with the entrepreneur-consumer relation. Also in this relation the consumer is generally conceived of as an intrinsically weaker party. At most, the discrepancy between the normative status and the factual condition of principals confirms the existence of a preliminary case in favour of regulation: there is a failure in the unregulated functioning of the considered institution.25 However, and this is the anomaly in our view, the mainstream microeconomic market rationale does not conceive of consumers as holding a normatively privileged position. b) When we turn to the question of whether the role of shareholders/investors is really so different from that of consumers/clients in their relationship to directors as the prime decision makers in firms, serious doubts arise. The core concept for making our case is consumer sovereignty. This concept implies a principal-agent relation between the consumer and the entrepreneur. Consumer sovereignty is seldom mentioned by mainstream law and economics. A notable exception is Parisi’s The Language of Law and Economics: “Austrian law and economics scholars use the term “consumer sovereignty” to refer to the principle that the consumer is the best judge of the types and quantities of goods that should be produced by the economic system. ... Many economists believe that consumer sovereignty could be undermined by several factors” generally identified as market failures.26 Two points of this definition are particularly important; one is historical, the other conceptual. Historically, this account of consumer sovereignty is too dismissive. In fact, it is true that the founding fathers of Austrian economics, von Mises and von Hayek, endorsed consumer sovereignty as foundational of their conception of market relations.27 23 Berle and Means (1991 (1932): 287); for both Manne (1963) and Berle and Means (1932), the context and later developments of their theories, see Grundmann (2015b: 1585-1587, 1598-1608 and 2015a: 1507-1517, 1526 et seq. respectively, forthcoming also in English). 24 From legal doctrinal theory: Grundmann (2010: 1071); from economic theory, path breaking Jensen and Meckling, (1976) and Alchian and Demsetz, (1972: 782 et seqq.) and also Berle and Means (1991 (1932): 112-116); for a recent discussion, see Buchanan, Heesang Chai and Deakin (2012: 40-61); for the development of principal agent theory and namely the context of the paper written by Jensen and Meckling, see Grundmann (2015a: 1507 et seq., 1517-1527, forthcoming also in English). 25 Indeed, the case is only preliminary because regulatory costs have to be taken into account. 26 Parisi (2013: 61). 27 Mises (1996 (1963): 269-273); Hayek (2014 (1948): 109; 1933: 214-217). 6 Department of Law Working Papers Investor-Consumer or Overall Welfare However, it is also true that modern “Austrian” economists are not so enthusiastic about it.28 Be this as it may for current “Austrian” economic thought, the most eminent protagonist of consumer sovereignty thinking, William Harold Hutt, did not see consumer sovereignty as minoritarian, but rather as foundational of economic thought29 – one could even speak of a “hidden mainstream”, as traces of this concept can be found in many works of truly foundational importance, even back to Adam Smith’s The Wealth of Nations.30 These traces are gathered elsewhere.31 Indeed, Parisi’s reference to the opinion of “many economists” connecting the limitation of consumer sovereignty to market failures should suffice as a preliminary confirmation of this claim. Conceptually, Parisi’s dictionary shows in a clear and concise way that consumer sovereignty challenges the foundational assumptions of mainstream economics of law. According to Parisi, in fact, “aggregate surplus is important in law and economics, in which economic and competition policies are designed to maximize aggregate surplus”.32 At the same time, consumer surplus “is at its highest, and the highest efficiency is attained, in perfectly competitive markets”.33 If we assume that market regulation and “economic and competition policies” are basically overlapping, the following can be observed: consumer sovereignty and the maximization of consumer surplus in perfect competition both point to the conclusion that efficiency is not about overall aggregate value but about consumer aggregate value. To the best of our knowledge, the author who has taken consumer sovereignty most seriously in the law and economics literature is Katalin Cseres. However, her understanding of the concept is reductive. Interestingly, the author presents consumer sovereignty as an element of mainstream theory. In her view, consumer sovereignty relates to the epistemic authority of the “Individuals [who] are assumed to be the best judges of their own welfare”.34 The normative law and economics standard she applies is nonetheless total welfare.35 In our reading of the literature, consumer sovereignty is a richer and more complex concept. At least two conceptions of consumer sovereignty can be found. In the autonomy conception, consumers have to exercise directly their sovereignty by demanding (or not) products.36 If the use of this sovereign power is detrimental to consumers, then there is a reason for limiting it. In the welfare conception of consumer sovereignty, instead, the consumer is sovereign also when public authorities are regulating market behaviour in order to increase consumer welfare – or in the consumers’ interest. The main problem of Parisi’s and Cseres’ conceptions of consumer sovereignty is not semantic (what are the elements of the concept), but pragmatic (what is the use of the concept). In fact, they do not raise consumer sovereignty to the status of a microeconomic market rationale, as the economic literature on the concept would suggest. The argument just sketched should be sufficient to accept consumer sovereignty as a challenger of the 28 The critique is built on the libertarian argument made by Rothbard (2004 (1973): 629-639). 29 Hutt (1940: 2): “fundamental notion … implicit, but seldom clearly stated, in applied economic writings in the orthodox tradition”. Hutt had already explained that “[h]is book is as much a criticism as a vindication of those who have been led to build on the traditions of orthodoxy” (Hutt (1936: 34)); in fact he considered to be the “main task of the economist or any other genuine social philosopher in th[o]se days … to determine the origin and then attack the false beliefs which actuate the society in which he finds himself” (Hutt (1936: 38)). See also Bymers (1982: 195), Reekie (1988: 6), Buchanan (1988: 6), Persky (1993: 184), Rutherford (2013: 113). Reekie further adds that “[b]y the mid-nineteenth century Smith’s views had prevailed” (Reekie (1988: 3)); De Strobel states that “[e]conomists have always stated that consumption is the ultimate end of production” (De Strobel (1970: 7, translated by Esposito; see also 21-28)). 30 In The Wealth of Nations, Smith writes that “Consumption is the sole end and purpose of all production; and the interest of the producer ought to be attended to, only so far as it may be necessary for promoting that of the consumer. The maxim is so perfectly self-evident, that it would be absurd to attempt to prove it” Smith (2007 (1776): 426). This is recognised, among others, in Iain Ramsay’s entry “consumer protection” in The New Palgrave Dictionary of Economics and the Law: “consumer protection reflected Smith’s statement that ‘consumption is the sole end and purpose of all production’” (Ramsay (2002: 410)). 31 Esposito (forthcoming: chapter 2). 32 Parisi (2013: 6). 33 Parisi (2013: 60). 34 Cseres (2005: 12). 35 Cseres (2005: 12, 176). 36 This conception is the one more in accord with Hutt’s original framework. European University Institute 7 Stefan Grundmann and Fabrizio Esposito mainstream approach based on overall value. Consumer sovereignty is a normative standard grounded in (micro)economic theory, after all! Why is the normative status of shareholders, and investors more generally, relevant for market analysis? Why, in other terms, should one care about the coherence of the analytical frameworks? One may observe that the market and the firm are two different institutions and, therefore, they can be governed by different normative standards. Against this view, there is a rather straightforward – indeed ‘logical’ and hence ‘formal’ – answer to the question just asked: the key idea is that the concept of investor – irrespective of whether the investment is made in shares or in bonds – is a particular instantiation of the concept of consumer. Hence, the principal agent relation accepted in mainstream corporate governance literature is a particular instantiation of consumer sovereignty – particular because related to the specific market of financial investment. The idea behind this claim is that each agent endowed with some wealth faces a choice: consume today or invest today in order to consume (hopefully more) tomorrow. Accordingly, the investor of today is the consumer of tomorrow. Also on the side of the instruments into which the consumer/investor ‘invests’ his money, there is this parallelism: the existence and function of the firm is explained in economic literature in terms of minimization of transaction costs and namely as an alternative to markets. The arrangement as a firm should be preferred if the advantages of the arrangement outweigh its disadvantages – with the advantages consisting in more stability and hierarchical command and search costs reduction and the disadvantages in agency costs, namely monitoring and screening costs and residual loss. Otherwise, market transactions should be preferred. This institutional comparison is clearly mainstream. Using the simple motto introduced by Williamson, the firm decides whether to “make or buy” with the aim of minimizing costs thereby increasing net returns – given the same value-output. Adding the premise that the directors are agents of the investors, we can conclude the higher the net returns, the higher the wealth available to investors. Therefore, the directors, acting as agents of the investors, (should) identify the mix of making and buying that maximizes the returns – and more generally, the interest – of investors. In light of the above, the principal conception of investors is an implication of the concept of consumer sovereignty. At the same time, the cost-minimization function of the firm is the counterpart of the price-minimization function of the market. A confirmation of the connection between the role of consumers in markets and investors in firms can be found in the diffusion of “exit” and “voice” as concepts of corporate governance. These expressions were introduced by Hirschman’s Voice, Exit, and Loyalty with regard to markets. In fact, Hirschman refers with “exit” to the market stimulus given by consumers moving to a different producer, while “voice” consists in consumers expressing dissatisfaction “directly to management or to some authority to which management is subordinate or through general protest”.37 Moreover, Hirschman relates market governance directly to consumer sovereignty when he argues that believing in “competition-exit … to solve most of the “sovereign” consumer’s problems” is an overstatement.38 The idea that this sovereignty/voice-exit conceptual apparatus has moved smoothly from the market literature to the corporation literature is inexplicable unless the role of consumers in markets is homologous to the role of investors in firms. At this point, it is important to remark the difference between this approach and the idea that market distribution does not matter because ‘we are all consumers’ (entrepreneurs on one market are consumers on another, so ultimately we are all consumers – the ‘Chicago trap’39) or ‘entrepreneurs are people too’ (there is no normative and, in particular, welfaristic reason to prefer consumers to entrepreneurs)40. We do not follow these arguments in holding that the role one plays on a market (consumer or entrepreneur) is not important to express a normative preference for a market allocation over another. To the contrary, we draw the attention to the fact that in corporate governance this normative hierarchy between roles has been an undisputed feature of mainstream thought for more than 80 years. And there are several connections between consumers and 37 Hirschman (1970: 4). 38 Hirschman (1970: 55); for Hirschman, the context and later developments of his theories, see Grundmann (2015b: 15851598, 1607 et seq.). 39 We borrow this expression from Hildebrand (2016: 30). 40 This version of the argument is used, for example, in Cowen and Tabarrok (2011: 225). 8 Department of Law Working Papers Investor-Consumer or Overall Welfare investors, markets and firms, as well as the relation of consumers with markets and investors with firms. We further observe that removing overall value from the microeconomic market rationale in favour of a normative preference for the consumer would solve this incoherence. c) Principal agent theory cannot only be the basis of this ‚logical‘ and ‚formal‘ parallelism between consumers/clients and investors/shareholders. The main substantive reasons given for postulating a hierarchy of interests enshrined in a principal agent model are worth consideration as well: they have been developed for equity holders,41 but they apply in a parallel way also to consumers/clients. The starting point of the argument is the relation between the normal concept of property and the choice of investing in a company. Investing in a company implies a delegation of powers over one’s own resources. Basically, you hire someone to manage your resources. The normative idea is simply that the property over the invested resources implies that, before the investment, those resources had to be used to maximize the interest of the owner (within the limits imposed by the law42). Thus, the delegation of powers over these resources does not change the function assigned to those resources in society: maximizing the interest of the (delegating) owner. Jensen/Meckling build on this normative idea to describe the functioning of the firm. For current purposes, they answer to the question “what are the corporate governance means for minimizing agency costs?” The famous answer and formula is: the non-managing investors need to monitor, the managers can give warranties (‚bonding expenses‘), but some remainder of diverted funds will always remain (‘residual loss’).43 The whole purpose of governance and covenants is to minimize these three types of costs – to render the firm as attractive as possible for non-managing investors. The market plays an important role of external governance because ultimately managers carry the costs of choosing the less efficient solution (p. 324 et seq.). In other words: the prime decision makers (managers) are motivated to find the agency costs minimizing arrangements. The procedural device to do so is the following: the management puts the complete arrangement on the table and beyond that looks solely at the benefit of the side of the capital market which (typically) does not initiate the arrangement. The external pressure exercised by potential managers and by alternative opportunities for actual and potential investors forces them to do so if they want to maintain their position and attract investors. The important caveat in Jensen/Meckling’s analysis is that capital markets are competitive, contracts are complete and investors are fully rational. From all this ensues what is the key finding for our purposes: shareholders are not conceived to be the principals because they are intrinsically weaker parties, in need of more protection (in fact, it may well be that they need less protection than managers at least in some respects, namely in that they can better diversify44). In fact, the core reason for conceiving shareholders and investors more generally as principals is rather that they own the resources managed by the company. We therefore call this “the property conception of the firm”. Can this line of argument be applied also to consumers? We claim it 41 Path breaking Berle and Means (1991 (1932)) and Jensen and Meckling, (1976); for the context and later developments of their theories, see Grundmann (2015a: 1507-1517, 1526 et seq. and 1507 et seq., 1517-1527 respectively, forthcoming also in English). 42 These limits are important because they embody considerations that are conflicting with allocative efficiency and that thus would require the legal system to balance the conflicting interests; see below, pp. 12-14. 43 Page 308: ‘If both parties to the relationship are utility maximizers there is good reason to believe that the agent will not always act in the best interests of the principal. The principal can limit divergences from his interests by establishing appropriate incentives for the agent and by incurring monitoring costs … [and may] pay the agent to expend resources (bonding costs) to guarantee that he will not take certain actions which would harm the principal …’. The overall scheme is easy to grasp as well: the management has the incentive now to overinvest in the consumption of a firm’s perquisites, as it profits from this (to a 100 %), yet no longer pays the expenses fully, but merely proportionally. Moreover, it may no longer invest so much effort into finding good opportunities, as it does no longer solely profit from them (according to Jensen/Meckling this is even the most serious risk, see p. 313). From the quote just made, the famous formula is derived that “agency costs”, i.e. costs of establishing an agency relationship (rather than a market relationship or assuming the position oneself) consist of the sum of: ‘(1) the monitoring expenditures by the principal, (2) the bonding expenditures by the agent, [and] (3) the residual loss.’ (also p. 308). See also p. 323-326. For agency costs – also as compared to (market) transaction costs – see also Fama and Jensen (1983); Williamson (1988); Hart (1995). 44 Note that this consideration holds on arguably heroic assumptions about the rationality and information of investors on the performance of the firm. European University Institute 9 Stefan Grundmann and Fabrizio Esposito can. The consumer has property over some resources that he can use directly for consumption or transfer to someone else in exchange for other resources. The choice is based on what increases more his own welfare. The act of transferring represents the homologous of the delegation of powers over property in a company. Entrepreneurs on the market – just like the management of a company – are best positioned to put an offer fully fleshed out on the table and for all the remainder look solely at the benefit of the other side. In this respect (but only in this respect), one may also argue that indeed investors/consumers are the weaker party – which is, however, just the other side of the coin: that mangers/entrepreneurs are and have to be more fully informed as they are invited to put the fully fleshed offers on the table. They do so driven by the invisible hand of the market – at least under the same caveat of perfect competition, contractual completeness and full rationality. In both cases, entrepreneurs are compelled by the market mechanism to work in the interest of someone else. Of course, if the market is not perfect, entrepreneurs will not be compelled to be perfect agents. Hence, also within this framework, the usual economic justification for legal intervention aimed at correcting market failures applies – but the argument made then brings us to argue that as well regulation of market failure has to follow this rationale, namely that consumer welfare needs to be the ultimate scope of such regulation. A confirmation of the soundness of this analysis can be found in the use of property language for blaming monopolists, but also company managers that are not optimal agents: they are all thieves! If they are thieves, then they are stealing someone’s else property. This language has been used, for example, by Willian Hutt with reference to monopolists (“monopoly … infuses into distribution an element of robbery”45) and by Judge Posner with reference to managerial overcompensation which has “redistributive effects [that] are obvious and troubling from an ethical standpoint because, by definition, overcompensation is a kind of theft from shareholders”.46 Posner’s statement is surprising: not only Posner takes part from the Chicago School’s positive attitude towards existing corporate management and finance structures47 but, more fundamentally, this “theft theory of harm” is basically incompatible with the idea that redistribution of wealth is irrelevant for economic analysis – an idea truly at the core of his wealth maximization framework and of his economic analysis of law.48 To the contrary, this theft theory is completely compatible with the conception of allocative efficiency suggested in this article. As stated above, we cannot claim victory on the grounds of this brief analysis. What we think it can comfortably be done is, instead, challenging the mainstream framework. In our understanding, that framework basically faces four alternatives. First, rejecting the conceptual connection between consumer and investor. In this way, the incoherence between investor supremacy in firms and the overall value standard in market analysis is avoided. If the connection is not rejected, the remaining three alternatives are indeed far reaching: one alternative is accepting overall value also in firm analysis, another is accepting consumer value also in market analysis, while the last alternative is accepting and ignore the incoherence in the theory. d) Before, in our second step of investigation, we take up the main reform measures in financial services and assess them from the perspective of client/consumer sovereignty by analysing them within a principal-multi agent framework, a word of caution is still required. It is about three qualifications on how far the paradigm of client/consumer welfare as the ultimate goal should reach. Three potential misunderstandings have to be addressed: the first is about the role of negotiations. The explanations made so far do in principle not limit the role of negotiation: when fixing interest rates for loans, banks do not need to charge the lowest possible rates – that is, they do not have the legal duty to minimize the interest rate in order to maximize consumer welfare. Limiting the amount of interest rates charged (or the burden stemming from other conditions) is the task of well-functioning, competitive markets: their ultimate microeconomic rationale is postulated to be (overall) consumer/client welfare and not overall welfare (the latter would include the benefit both of consumers/clients and of entrepreneurs). Outside 45 Hutt (1936: 99-100). 46 Posner (2009: 1042). Indeed, Posner is concerned also with social costs, but both “redistributive effects and social costs … appear sufficiently large … to warrant serious consideration of possible ameliorative measures“ (Posner 2009: 1043). 47 See Medema (2010) for a discussion. 48 Posner (1985). 10 Department of Law Working Papers Investor-Consumer or Overall Welfare situations of market failure (including the behavioural ones), each side takes care of her interests. In fact, it falls within the institutional prerogatives of well-functioning markets to foster consumer/client welfare through competition. It is only in situations where the supplier/manager is making decision unilaterally – as an agent – but without adequate incentives to act in the interest of the principal that the welfare of the latter serves as the normative guideline for market behaviour regulation. In synthesis, we do not argue that legal intervention has to substitute the market mechanism, it has to support it. So far, then, our view is conceptually entirely consistent with the mainstream position at the prescriptive level. The second qualification is about the role of entrepreneurs’ welfare or interest. The focus on client/consumer welfare maximization does not imply that entrepreneurs’ welfare is irrelevant. To put it differently, entrepreneurs cannot be exploited, welfare cannot be sucked out of them indefinitely in order to improve (even slightly) consumer welfare. As already stated, in our understanding of allocative efficiency there is the constraint of mutual Pareto optimal transactions. In the terms of principal-agent theory, while the client/consumer welfare is the maximand, there are a participation and an incentive constrains to be satisfied.49 This qualification is normatively important and our experience of discussing this research with others suggests that the concern is immediate and powerful. However, in our view, this constraint is already enshrined in the first qualification. In fact, a necessary element of a wellfunctioning, competitive market is the freedom to choose whether to contract and if so, with whom. 50 Therefore, given that the entrepreneur enjoys these freedoms, he will operate on the market only if this makes him better-off. Thus, the market process embodies a certain level of protection of entrepreneurs’ welfare that we have expressed with the prerequisite of mutual Pareto optimal transactions. From a different perspective, a Pareto optimal market equilibrium is allocatively optimal only when consumer welfare is maximized. It follows that as long as the agency costs imposed on the consumer are not minimized, the resulting equilibrium cannot represent the ideal allocative optimum. This is the point where our position departs from the current mainstream. The third qualification relates to the incompleteness of the interests considered so far. Indeed, already in welfare economics the Second Welfare Theorem reminds us that a just society cannot be reduced to an allocatively optimal general equilibrium. When it is postulated that the ultimate microeconomic rationale of markets or of market regulation – just like the ultimate goal of company law regulation – is overall client/consumer welfare (or shareholder/investor welfare) via mutually Pareto optimal transactions, this relates to the relationship between the classes of actors named: shareholders/investors and managers on the one hand and clients/consumers and entrepreneurs on the other. What it does not relate to is collective and third party interests (externalities51): just as in company law, for instance, regulation can protect the interest of overall financial stability or economic growth irrespective of the fact that managers are seen as agents and shareholders as principals, third party and public interests are to be considered separately from the relationship between entrepreneurs (as agents) and clients/consumers (as principals). Consumer sovereignty/empowerment as terms refer again solely to the relationship between suppliers and consumers. This standard does not go any further than considering the relation between consumers – in the sense that the costs for non-vulnerable consumers of protecting vulnerable consumers need to be considered in the optimization calculus. One could say that distinguishing a micro (allocative efficiency) and a macro (growth) market rationale already shows that our normative framework does not reduce a just society to an allocatively optimal general equilibrium. However, even considering growth is not enough. Indeed, in a just society the distribution of resources and arguments about responsibility matter too.52 49 See below, pp. 14-18. 50 For a legal discussion along these lines, see Study Group on a European Civil Code and the Research Group on EC Private Law (Acquis Group) (2009: 59); from an economic perspective, see Stigler (2008). 51 Our characterization of the other relevant interests here is admittedly just a sketch. However, for current purposes it appears to be sufficient. 52 As far as we know, the most elaborate study in this direction in the law and economics literature was offered by Adler (2012). European University Institute 11 Stefan Grundmann and Fabrizio Esposito Regulated market behaviour and allocative efficiency: a principal multi agent framework The previous section has advanced a normative framework in which the microeconomic market rationale (allocative efficiency) is defined as aggregate consumer welfare maximization via mutual Pareto optimal transactions. The framework then conceives of market relations as principal-agent relations between consumers and entrepreneurs. The economic pedigree of this framework was identified primarily in the consumer sovereignty literature, but emphasis was put also on some of the conceptual difficulties incurred when this perspective is denied. In particular, denying the normative symmetry between firm and market relations causes a conceptual incoherence that is particularly disturbing given the foundation of the choice between consumption and investment in an allocation of one’s own economic power granted by property. Moreover, a foundational distinction in the corporate governance literature exitvoice (internal-external governance) was originally conceptualized with reference to the consumerentrepreneur relation. In the analysis in the previous section, the law has remained mainly on the background, as the device allowing parties to design their principal-agent relations through contract law. It is now necessary to enrich this framework: this view of the role of the law as a neutral and passive tool is too reductive and simplistic. In this section, we take legal mechanism more seriously. To do so, we first discuss different conceptions of a principal-agent interaction and we explain why we choose to base our analysis on an account in which the principal is the economic actor delegating a choice to another actor and the maximizand is the interest of the principal. Then, we look at the way in which a principal-multi agent framework that takes into account the interplay between consumers/investors and entrepreneurs/managers as well as the legal system can look like. Finally, we show that the elements of this principal-multi agent framework are present in the foundations of EU consumer and corporate policies. Justifying the delegation model in principal-agent theory The 2016 “Nobel” Prize in Economics has been awarded to Oliver Hart and Bengt Holmström “for their contribution to contract theory”. The core problem investigated by the awarded economists is the study of what happens when “A principal engages an agent to take certain actions on the principal’s behalf”.53 As Laffont and Martimor explain in their influential The Theory of Incentives, principal-agent theory focuses primarily on two information problems principals can have.54 First, “the agent can take an action unobserved by the principal, the case of moral hazard or hidden action”. Second, “the agent has some private information about its cost or valuation that is ignored by the principal, the case of adverse selection or hidden knowledge”. A third information problem is the non-verifiability by a third party, in particular by a “Court of Justice”, of information shared by the principal and the agent. The aim of the inquiry is to “stud(y) when this private information [of the agent] is a problem for the principal, and what is the optimal way to cope with it”. A principal-agent interaction can be represented as a system of equations. Taking shareholders as the principal and the managers as the agent, Aglietta and Rebérioux model the interaction as follows:55 In this model, V(w) is the value for the principal of the effort e of the agent. The agent can choose between two levels of effort e+ and e- and receives a wage w. Assume that V(w) is maximized if the 53 The Committee for the Prize in Economic Science in Memory of Alfred Nobel (2016: 1). 54 Laffont and Martimor (2002). 55 Aglietta and Rebérioux (2005: 30). We have chosen this representation for its analytical clarity for current purposes. 12 Department of Law Working Papers Investor-Consumer or Overall Welfare agents takes action e+ instead of e-. The model then imposes two feasibility constraints—the participation and the incentive constraints—that will be sufficient to ensure that e+ is taken, thereby maximizing V(w). The participation constrain posits that the utility for the agent of e+ is higher than the utility of any other possible contract available on the market to the agent. The incentive constrain posits that the agent gains more utility by action e+ than by action e-. When these two constrains are satisfied, it is in the interest of the agent to act in the interest of the principal—that is, to maximize V(w). Thus, the goal—the maximization of the principal’s interest—is achieved spontaneously because action e+ is Pareto superior to action e- in the strongest possible sense, namely, all economic actors are better off. Laffont and Martimor have a similar account, but they express the principal’s objective function as a trade-off between allocative efficiency and information rent.56 Under this account, it is optimal for the principal to increase the information rent as long as this implies a superior increase in expected allocative efficiency. In other terms, expected allocative efficiency represent the benefit for the principal and information rents the cost that the principal has to transfer to the agent to achieve the benefit. In both accounts of a principal-agent interaction, the optimal condition is the one that makes the principal as well off as possible while making the agent willing to cooperate. However, in a footnote Laffont and Martimor observe that “a social utility maximizer putting equal weight on the principal and the agent’s expected utility in his objective function would be interested in maximizing expected allocative efficiency only, without any concern for the distribution of information rents between the principal and the agent”.57 We have just encountered the first ambiguity in the principal-agent relation. Whose value, welfare, interest is the interaction instrumental to? The value, welfare, interest of the principal, or that identified by the social planner? The Nobel Prize Committee takes the latter view, but then is very careful in pointing out that with transfers between the principal and the agent “any desired distribution of surplus between [them] can be achieved”.58 Similarly, in a 2001 article, Hart identifies the interest of principal, of the agent and then adds that it is “useful to write down the objective of a planner who is concerned with social (or Pareto) efficiency” and observes “that these three objective functions are generally distinct”.59 The question then, in our view, is the following: provided that both the principal and the social planner view are used in economic theory, which one is the one more in accord with legal practice? This is the first question a law and economics approach should answer.60 The problem is even more complex once you realize that there is a different and overlapping ambiguity regarding the criterion for applying the principal-agent framework, namely that between the delegation criterion and the information problem criterion. According to the delegation criterion, the principal is the economic actor delegating an action to another economic actor, the agent. According to the information problem criterion, the principal is the economic actor facing a problem of hidden information or action on the side of the other economic actor, the agent. The delegation criterion is the one taken by the Nobel Prize Committee when they state that the agent “takes an action on the principal’s behalf”. Similarly, Laffont and Martimor open their analysis of the “rent extraction-efficiency trade-off” observing that:61 Incentive problems arise when a principal wants to delegate a task to an agent. Delegation can be motivated either by the possibility of benefitting from some increasing returns associated with the 56 Laffont and Martimor (2002: especially chapter 2). Note also that their simplest model is already far more sophisticated than the one presented by Aglietta and Rebérioux. 57 Laffont and Martimor (2002: 48). 58 The Committee for the Prize in Economic Science in Memory of Alfred Nobel (2016: 6). 59 Hart (2001: 1086). 60 See below, pp. 38-40. 61 Laffont and Martimor (2002: 37). European University Institute 13 Stefan Grundmann and Fabrizio Esposito division of tasks which is at the root of economic progress, by the principal’s lack of time or lack of any ability to perform the task himself, and, finally, by any other form of the principal’s bounded rationality when facing complex problems. The delegation criterion is compatible with the principal’s welfare criterion under the observation that principals delegate in their own interest; therefore, it makes sense to identify the delegator with the economic actor – namely the principal – whose interest is prioritized. Moreover, the idea of delegation is at the core of the definition of consumer sovereignty given by William Hutt. For Hutt:62 The consumer is sovereign when, in his role of citizen, he has not delegated to political institutions for authoritarian use the power which he can exercise socially through his power to demand (or refrain from demanding). Thus, the coherency argument developed above supports the delegation criterion. Unfortunately, both the Nobel Prize Committee63 and Laffont and Martimor use also the information problem criterion. For example, Laffont and Martimor state that “When the monopoly has private information about its cost or demand, its regulation by a commission becomes a principal-agent problem” or that “Exchange raises incentive issues when the commodity which is bought has a value unknown to the buyer but known to the seller”.64 In the framework we introduce in the next section, the former is a problem of teamwork between a primary agent—the entrepreneur—and the monitoring agent—the legal system.65 To the contrary, the second is a problem of teamwork between the primary agents that should be solved maximizing the value of the exchange for the buyer. What is, in our view, puzzling is what example Laffont and Martimor give of principal-agent problems between an uninformed buyer and an informed seller:66 It is the case, in particular, in insurance markets when the insurance company buys a risk plagued by moral hazard or adverse selection. The insurance company faces a principal-agent problem with each insured agent, but may nevertheless have a statistical knowledge of the distribution of risks. What we find odd is the idea that the insurer buys the risk and, therefore, the insured sells the risk. An account of the interaction that is more in accord with the common and the legal understanding of an insurance contract is that the insurer is a service provider and the insured buys the service and pays for it. We do not deny that the insurer has hidden information problems. We deny, however, that this is a reason for turning the insurer into a principal. Simply, the insurer is an agent facing hidden information problems. Indeed, the insurance contract is exceptional among contracts insofar as the problem of hidden information is heavier for the service provider than for the client – which is typically the other way around (information on the quality of offers and services, but also of the match with clients’ needs being typically more easily available to the provider). This exceptional distribution of information problems in insurance contracts, however, does not change the basic distribution of roles, i.e. that the service provider formulates the fully fleshed offers from which clients choose. And if this sounds odd to you, be reminded that this is the way in which Arrow framed the hidden information problem in his groundbreaking article Uncertainty and the Welfare Economics of Medical Care: for Arrow, “moral hazard” is the problem of “the effect of insurance on incentives”; nevertheless, Arrow qualifies insurance companies as agents.67 Moreover, if we go back to the account of the delegation criterion by Laffont and Martimor, we clearly see that the hidden information or action problem is not a necessary condition for 62 Hutt (1936: 257). 63 The Committee for the Prize in Economic Science in Memory of Alfred Nobel (2016: 1) consider from the beginning a delegating principal who also “cannot directly observe the agent’s actions”. 64 Laffont and Martimor (2002: 17-18). 65 See below, pp. 18-21. 66 Laffont and Martimor (2002: 18). 67 Arrow (1963: moral hazard 961; insurance company as agent 960). 14 Department of Law Working Papers Investor-Consumer or Overall Welfare the application of the delegation criterion. Similarly, if we go back to the model by Aglietta and Rebérioux, we see that—also there—no reference to any hidden information or action is made. To sum up, one can have four different conceptions of a principal-agent interaction: 1. Delegating principal in his own interest 2. Delegating principal with the interest to be maximized chosen by the social planner 3. Uninformed principal whose interest has to be maximized 4. Uninformed principal with the interest to be maximized chosen by the social planner In our view, the more you move from i. to iv., the more obscure it becomes why an economic actor is designed as principal in the first place. In i., there is a perfect correlation between the normative idea of a delegation of actions (tasks, powers, responsibilities, decisions, etc.) and the interest to be pursued. Note that the model by Aglietta and Rebérioux includes only constraints necessary to make the agent willing to cooperate to the principal’s plan, but it could be easily extended to consider, for example, negative externalities or other desirable constraints for the social planner. This circumstance is sufficient to satisfy the constraint on our consumer-principal view discussed at pp. 7-12. In ii., you already see that the social planner takes normative prominence over the delegating principal. In these circumstances, it might well make more sense to consider both actors as agents and the social planner as the principal whose plan has to be followed. In iii., the reason why there is a normative connection between information problems and the interest to be maximized is unclear. Take an entrepreneur that does not know the reserve price of its consumers. Are the consumers the agents of the entrepreneur because there is a problem of hidden information and, in turn, this justifies the normative claim that the interest of the entrepreneur ought to be maximized? This account of the interaction is clearly odd. Finally, in iv. there is not only the problem identified in ii., but also the further problem that why the uninformed actor, whose interest is not even the maximand, deserves to have such a focal label as “principal” is unclear. The label “principal” suggests that this economic actor is in some sense central in the interaction—when clearly everything that matters both at the normative and at the instrumental levels lays somewhere else: the goal is set by the social planner and the important actions are taken by the agent. Three final observations are relevant here. First, since we find the delegating principal’s welfare conception of principal-agent theory the most convincing, this is the conception we will use from now on. Second, we emphasise that it is a possible economic conception of principal-agent theory. It cannot be dismissed by an economist as a non-economic conception. Third, we have to rule out a further possible account of the relation between the principal’s interest and the social planner’s view. According to this account, pursuing the interest of the principal is instrumental to pursue total welfare—which, as seen above, is a typical account of the social planner’s view. In other terms, the interest of the principal is just a means to total welfare. This claim has been clearly advanced in the economic analysis of corporate law by influential authors like Hansmann and Kraakman. In their view, in 2001 corporate law had reached the end of history because “There is no longer any serious competitor to the view that corporate law should principally strive to increase long-term shareholder value”.68 In their view, the “shareholder-oriented model” has prevailed because of its “efficiencies”. In the Anatomy of Corporate Law they develop the point as follows: “the appropriate goal of corporate law is to advance the aggregate welfare of all who are affected by a firm’s activities … . This is what economists would characterize as the pursuit of overall social welfare”.69 In fact, if we take the “shareholder-oriented model”—as they called it in 2001—at “face value” it is neither descriptive nor “a normatively appealing aspiration for that body of law”: it would imply that you can make the workers worse off of 2 euro to make the shareholders better off of 1 euro. Instead, they continue, the shareholder-oriented model: 68 Hansmann and Kraakman (2001: 439). 69 Armour, Hansmann and Kraakman (2009: par. 1.5) Note that there are some differences in the way the argument is made in the previous edition (Hansmann and Kraakman 2004: 18-19). European University Institute 15 Stefan Grundmann and Fabrizio Esposito can be understood as saying, more modestly, that focusing principally on the maximization of shareholder return is, in general, the best means by which corporate law can serve the broader goal of advancing overall social welfare … We believe that this second view is—and surely ought to be—the appropriate interpretation [of the shareholder-oriented model]. The instrumentalization of shareholder value to the goal of overall social welfare rests on the observation that shareholders will make sure that interactions will be beneficial also for “creditors, workers, and customers” in order to obtain their consent. No further elaboration of the point is given, but the authors conclude the section by observing that “each of the authors of th(e) book has individual views on” the effectiveness of this instrumentalization and “no strong position” is taken on the point. Instead, they merely intend to offer “an analytic framework”. Indeed, the instrumentalization argument is not very convincing—not even for the arguers. And, in fact, if we reconsider our discussion of the model by Aglietta and Rebérioux, one can see that making the agents better off in a principal-agent framework can be easily turned into a means to shareholder’s value. To what extent one account is morally more plausible than the other is beyond the interest of this article.70 What we care about is the extent to which the distributive concern between the principal and the agent is relevant in the law. If it is, as we argue, then also economists have a reason for choosing the delegating principal conception of the principalagent model. This is a possible economic conception of a principal-agent problem and, if one is convinced by our argument, it is more in accord with legal practice than the alternatives. Ultimately, leaving to legal institutions the task of answering normative questions in case of doubt is arguably consistent with the idea of economics as a value-free science. On the grounds of these clarifications, we can now turn to the construction of our principal multi agent framework. Building the principal multi agent framework A principal multi agent model describes different modes of interaction between agents and ranks them as means to the end of fostering the interest of the principal. Quite interestingly, Holmström’s foundational contribution to this literature was inspired by the research in corporate governance assuming the property conception of the firm.71 Agents constitute a team and we refer to their actions as teamwork. Unfortunately, the standard economic analyses of this topic do not seem particularly useful for current purposes.72 The agents generally have homogeneous roles; the strategies are simple, abstract and categorical; the focus is on agency costs. Nevertheless, if we consider the basic information problems identified by Laffont and Martimor, we see already that the only awareness of the problem of nonverifiability by “Courts of Justice” opens the way to conceive more accurately the role of the legal system in a principal-agent interaction. Hoping for more granularity and interdisciplinary dialogue73 in the future, in the meantime we looked somewhere else. We build on what we consider a truly seminal article by Clayton Gillette, for current purposes but perhaps even beyond, and on a more recent contribution, notably co-authored by two of the most prominent figures in the current American law and economics literature, Cass Sunstein and Oren Bar-Gill. Gillette, in particular, offers an analysis of the interplay between the different agents available in a regulated market process namely consumers, entrepreneurs, and the legal system. However, the most interesting part of his work consists in a nuanced reconstruction of the teamwork of the primary agents. Sunstein and Bar-Gill, instead, accept the framework developed by Gillette, but are actually more interested in the interplay between the single individual and the legal system. In fact, they offer a general conception of regulation as delegation, in 70 On the point, see for example the discussions in Aglietta and Rebéroux (2005) and Heath (2014). 71 Holmström (1982). 72 See, for example, Bolton and Dewatripont (2005: 297-364). Unfortunately, Laffont and Martimor never published the second volume of The Theory of Incentives, which was planned to analyze the problem of teamwork. 73 We cannot in fact exclude that more useful economic modelling of principal and multi agent relations are available and would be thankful for suggestions. 16 Department of Law Working Papers Investor-Consumer or Overall Welfare which not only the legal system acts as agent of individual principals, but can also enlist individuals as agents of its own agenda (“reverse regulation”). In what follows, we start with Sunstein and Bar-Gill’s framework and then move to Gillette’s. The reason is that Sunstein and Bar-Gill’s framework is much more detailed from the perspective we are more interested in, namely, the role played by regulation in our principal multi agent framework. Note that the purpose is purely analytical. We have no interest here in discussing the normative desirability of the different forms of teamwork. What we are interested in is looking at the aforementioned policies and directives in this and the next sections on the basis of a principled framework that allows to qualify different forms of intervention as expression of a principal multi agent approach, with consumers as principals. a) Sunstein and Bar-Gill’s focus on regulation suggests that an interesting way to reconstruct this three agents model is that of considering consumers and entrepreneurs as primary agents and the legal system as monitoring agent of the primary agents.74 In this way, our approach is practically a different reconstruction of the correcting market failures approach in which the legal system intervenes in market relations to the extent these relations are, to some extent, defective. As a monitoring agent, the legal system has six strategies on a spectrum from no delegation to full delegation. In case of no delegation, the legal system remains on the background, giving full enforcement to whatever transaction results from the teamwork of primary agents. In case of full delegation, the legal system determines unilaterally, via mandatory rules, one or more attributes of the transaction. Importantly, both categories are actually more theoretical than practical. In practice, some form of intrusion in the teamwork of primary agents always exists – at a minimal level with lack of consent defences. In these cases, some form of monitoring exists. The situation is even more complex when one considers full delegation more closely. First, mandatory rules often ban certain conducts rather than establishing what conduct is required. It is not by chance that EU law speaks of “unfair terms regulation” instead of “fair terms imposition”. Second, primary agents are generally involved in the legislative process as stakeholders, so that no mandatory rule is created without some form of teamwork with the primary agents. Third, legal provisions are always uncertain, to some extent. Hence, when a legal provision is invoked as grounds for requiring a behaviour by a primary agent, some teamwork – negotiations, out of court settlements, litigation – is needed to identify this behaviour. Within the spectrum, Sunstein and Bar-Gill distinguish information delegation, veto-based delegation, incentive-based delegation and commitment delegation.75 With information delegation, the legal system makes sure that some information is shared between the primary agents. Typically, the entrepreneur has to inform the consumer, but it is also possible that the consumer has to inform the entrepreneur – in our context, the most important example of the latter case is enshrined in the duty of responsible borrowing.76 Veto-based delegation focuses on the use of default terms. Unless primary agents decide otherwise, the default becomes part of the contractual terms. With incentive-based delegation, the legal system influences the behaviour of the primary agents with money. Liability rules work this way. But also with administrative sanctions or tax benefits the legal system gives incentives without mandating an outcome. With commitment delegation, the legal system helps primary agents to behave consistently with previous commitments. The authors give as example the famous 401(k) retirement saving system. This category shows how important it is to keep explicit – contrary to what Sunstein and Bar-Gill do – the role of the legal system as monitoring agent of primary agents. In fact, once the legal system is a monitoring agent, we see that commitment delegation is not the most intrusive form of delegation after full delegation. Instead, it is actually the mechanism by which the legal system exercises its monitoring function in case of no delegation. From a similar perspective, it is interesting to focus on the idea that information disclosure is less intrusive than veto-based disclosure. While this is probably true for the consumer, it is actually the contrary for the entrepreneur. In fact, while information disclosure is 74 Varian (1990). 75 Bar-Gill and Sunstein (2015: 15-19). 76 See below, pp. 36-37. European University Institute 17 Stefan Grundmann and Fabrizio Esposito mandatory for the entrepreneur, in veto-based delegation the veto can be easily exercised via standard terms. We think it is quite telling of the deep normative foundations of their framework that Bar-Gill and Sunstein implicitly adopt the point of view of the consumer. b) While Sunstein and Bar-Gill focus mainly on the monitoring function of the legal system, Gillette is more interested in the teamwork of primary agents. It is possible to identify three different models of teamwork: the moderately self-interested entrepreneur model; the consumer screening model; and the competition for active consumers model.77 The labels are ours. In the moderately self-interested seller model, quite simply, the entrepreneur acts in the interest of the consumer because he prefers to treat the consumer fairly, at least to some extent. The role of the consumer as agent is secondary. In the consumer screening type of teamwork, the entrepreneur apparently acts as a bad agent of consumers generally. For example, by introducing standard terms that are egregiously one-sided. In truth, however, he is going to enforce those terms only against consumers that abuse of contractual rights – for example, consumers claiming a product to be defective when they misused it or using a cooling-off period as a free trial. Thus, the real bad agent are other consumers, and the consumer screening agent acts in order to remedy to the difficulty of the legal system to monitor the behaviour of these bad agents. In case of competition for active consumers, entrepreneurs compete with each other in order to contract with active consumers, ending up fostering the consumer interest generally. Here, some consumers need to be active – they have to switch from one operator to the other in search of better bargains and/or they have at least to read standard contract terms to improve their quality. In Hirschman’s language, active consumers exercise exit. In Hutt’s language, active consumers are the ones exercising the power of demanding or refraining from demanding. As already stated, we do not discuss the desirability of different types of teamwork. Being this inquiry so empirically driven, it is even debatable whether lawyers are actually fit for this task or they should rather ask for and build on the work of scholars from other disciplines. Before moving to the application of the framework, it is important to emphasise the advantage of our framework against the ones just discussed. c) To the aforementioned analyses, we add a clear normative foundation in economic theory of the principal-agent framework, namely the claim that the microeconomic rationale of market behaviour and of its regulation is concerned with aggregate consumer welfare through Pareto optimal transactions. Both analytical frameworks are defended by their authors on mere theoretical grounds. They offer interesting perspectives on existing legal debates.78 No effort is, however, made to justify normatively the qualification of the consumer as principal. Actually, from a normative perspective, these approaches are dissatisfactory. They are, in fact, contradictory. Gillette primarily refers to the internalization of the principal’s interest. However, this internalization is associated with no less than five different normative standards: preference satisfaction (p. 685); best internalization of the principal’s interest (p. 689); ensuring mutual gains (p. 690); total welfare (p. 702); fair distribution of gains (p. 712). First of all, preferences are not a conclusive indicator of welfare, as they can be uninformed or non-fully rational. The implication is the need of a more elaborate notion of welfare. Setting this problem aside, only the best internalization standard is fully compatible with a principal-agent approach. Ensuring a fair distribution of gains goes in this direction, because generally the distributive problem is that contracts are unbalanced in favour of the entrepreneur, but a fair distribution could allow for a sub-optimal implementation of the principal’s interest as long as this outcome is deemed fair. The mutual gains and the total welfare standard are even more problematic. Let us set aside the conflict between these two criteria – the case in which one party is harmed but total welfare is maximized. According to both standards, there is only lip service to the idea of the consumer as principal. In fact, in the mutual gains standard, it is clear that the consumer is put on equal grounds with the entrepreneur and neither maximization nor distribution of gains matters. All that matters is that both parties are made better off. Even worse for the total welfare standard. In this case, the principal’s interest is merely instrumental (indeed, the same goes for the interest of the entrepreneur) towards an aggregate goal – total welfare – that admits the sacrifice of the “principal’s” interest as long as the aggregate value increases. 77 Gillette (2004: 690-712). 78 Gillette (2004: 689-690); Bar-Gill and Sunstein (2015: 1-4). 18 Department of Law Working Papers Investor-Consumer or Overall Welfare Unsurprisingly, these considerations were already made with regard to the social planner model of principal-agent relation (pp. 15-18). Sunstein and Bar-Gill, instead, originally declare commitment to social welfare analysis, and argue that the principal agent framework can actually be used for “pushing welfare analysis further and deeper” (p. 3). However, when they apply the framework to regulated market behaviour, the connection with social welfare becomes mysterious: when consumers are not rational, if the legal system does not act as consumer-agent, entrepreneurs can “steer consumers towards products that are less beneficial (to consumers) but more profitable (to sellers)” (p. 14). If their approach aims at social welfare maximization (as they claim), then there is no problem in reducing benefits to consumers to increase profits for entrepreneurs as long as the benefits offset the losses. But, again, if this is the case, total welfare is the principal, not the consumer. This confusion is to some extent typical of the economic analysis of contract law. 79 It is even less surprising once the mainstream approach meets principal-agent theory and qualifies the consumer as principal. In fact, assuming the total welfare standard and the principal-agent approach while qualifying the consumer as principal, determines a normative conflict in all those cases in which increasing total welfare is detrimental to the consumer-“principal”. In fact, from an overall welfare perspective – under the heroic assumption of the welfare neutrality of distributive effects – one cares only about the deadweight loss, which measures the reduction in total welfare. To the contrary, in a principal-agent model, the principal is the residual claimant. In this framework, the homologous of the deadweight loss is the residual loss. Both concepts measure the difference between optimal conditions and possible outcomes. However, the residual loss is the reduction in the welfare of the principal that remains (is residual) when agency costs are minimized.80 As the minimization of agency costs advantages the principal and does not necessarily increase overall welfare, the connection between the two standards is purely contingent. Our conception of allocative efficiency offers a simple way out. Total value does not matter. Only consumer welfare has intrinsic value and should be maximized. Then, the welfare of the entrepreneur matters to the extent that increasing it is instrumental to ensuring the agent’s cooperation towards more allocatively efficient market outcomes. Just like the welfare of agents in a principal-agent model is increased only to the extent that this is necessary to improve the outcome for the principal. The principal-agent structure of EU consumer and corporate governance policies In this section, we discuss the basic features of the EU consumer and corporate governance policies. The aim is to show that our principal-multi agent framework applies equally well to both cases when the focus is on the microeconomic dimension of the policy. When, instead, the macro dimension becomes also explicit, the framework becomes more complex because of the interaction between the micro and the macro market rationales. We focus on EU consumer and corporate governance policies because while we argue they are conceptually connected, they are generally seen as disconnected from a legal-systematic point of view. Thus, coherence here confirms the soundness of the framework sketched at pp. 6-14, but it also gives a first example of its explanatory power. Moreover, even though they are directly disconnected from a (mainstream) legal-systematic point of view, indirectly they are connected by financial services – the topic of the next three sections of this article. In fact, it is undeniable that in financial services regulation there is both a consumer and a corporate governance dimension. 79 See Esposito (2016) for a discussion of the tension between the Pareto criterion, the Kaldor-Hicks criterion and allocative efficiency as defined in this article. See also Esposito (forthcoming; chapters 1 and 4). 80 See Jensen and Meckling (1976: 308). European University Institute 19 Stefan Grundmann and Fabrizio Esposito a) Consumer policy has two pillars, consumer empowerment and consumer protection.81 Consumer lawyers typically associate the idea of consumer empowerment to the information duties of entrepreneurs in favour of consumers.82 Indeed, in the most recent Communications by the European Commission on its consumer policy (2007, 2012), this is the way in which the expression is primarily used. Importantly, the policy goal is to have consumers in the “driving seats” of the economy83 – arguably a different way to express the idea of consumer sovereignty. Moreover, we also see that empowerment aims at enhancing the competition for active consumers forms of teamwork of primary agents. Normatively, empowerment clearly counts as a procedural standard instrumental to a substantive goal: “Final outcomes for consumers in economic and non-economic terms are the ultimate arbiter of whether markets are failing or succeeding in terms of citizens’ expectations”.84 The improvements can consist in “greater choice, lower prices, and the affordability and availability of essential services”. 85 Interestingly, these are all benefits that a perfectly competitive (hence, allocatively efficient) market would achieve – with the notable exception of the availability of essential services, which opens the way to the concept of access justice.86 In other terms, the economic interest of the consumer is basically defined by reference to the outcome of perfect competition. The Commission however uses also another notion of consumer empowerment, which relates to consumers participation in the political process – a notion, however, which falls outside the conceptions of consumer sovereignty named and discussed above.87 The idea of consumer empowerment thus includes, but is also broader than, the autonomy conception of consumer sovereignty. First, giving information and educating consumers – “helping consumers to help themselves”88 – is part of the idea of empowerment. The goal is having consumers that make “rational decisions”.89 From this perspective, the EU information paradigm is somewhere in between the autonomy and the welfare conception of consumer sovereignty: the legal system intervenes, but the intervention is merely procedural (perhaps, it is a form of means-paternalism90). Second, the idea of empowerment is also associated with the participation of consumers in the legislative process as stakeholders. Clearly, participating in the political process counts as exercising sovereignty as citizens and, thus, there is that element of delegation of decisional power that is incompatible with the autonomy conception of consumer sovereignty. As the welfare conception is about the outcomes, this political idea of empowerment is also alien to the welfare conception of consumer sovereignty. Nonetheless, the abovementioned considerations about regulation as a form of teamwork between the legal system and the primary agents of allocative efficiency allow to account for this role of consumers as stakeholders. Consumer protection refers to those cases where consumers cannot help themselves. 91 In other terms, empowering consumers is considered an insufficient form of teamwork to foster allocative efficiency. For example, the intervention in health and safety issues is justified by the idea that there are “risks and 81 European Commission (2007: 10). The distinction between empowerment and protection is used also by scholars; see, for example, Reich (2016: 152); Zuiderveen Borgesious (2015: 198). Cseres (2005: 320-326) refers to empowerment as “liberal model” and protection as “paternalistic model”; however, the author conceives these as alternative goals of consumer protection, while we submit that it is more appropriate to consider them as two means of consumer policy. 82 See, for example, Weatherhill (2016) and Mak (2011). 83 European Commission (2007: 5 and also 2). See also European Commission (2012: 2), where it is stated that "Empowering consumers means providing a robust framework of principles and tools that enable them to drive a smart, sustainable and inclusive economy. Empowered consumers who can rely on a robust framework ensuring their safety, information, education, rights, means of redress and enforcement, can actively participate in the market and make it work for them by exercising their power of choice and by having their rights properly enforced” (emphasis added). 84 European Commission (2007: 3). 85 European Commission (2002: C 137/3). See also European Commission (2005: 9). 86 The concept has been pioneered by Hans-Wolfgang Micklitz in Micklitz (2011: 5, 34-43). 87 European Commission (1999: 2; 2002: 2). 88 European Commission (1999: 10). 89 European Commission (2005: 45; 2007: 4, 9). 90 It is complex to decide whether regulating conflict of interests between two parties mandatorily in market behaviour is a paternalistic move. For example, Cseres (2005: 320-321) and Sibony and Helleringer (2015) adopt this view. 91 European Commission (2005: 3; 2007: 5). 20 Department of Law Working Papers Investor-Consumer or Overall Welfare threats which are beyond the control of individuals”.92 Already information economics explained that the costs of controlling the quality of certain attributes related to safety and health are so high that it is better if the law regulates them.93 Notably, an important sub-set of consumer protection rules relates to vulnerable consumers. In case of vulnerability, not all consumers are equally empowered. The idea then is that enhancing competition for the empowered consumers is not an allocatively efficient form of teamwork. In general, then, when protection is given, the legal system intervenes intrusively in the teamwork of the primary agents. Notably, this approach falls beyond the scope of the autonomy conception of consumer sovereignty. However, it is fully consistent with the welfare conception of this economic concept. The results of this conceptual comparison are important. Considering the welfare of consumers as the rationale of consumer policy is fully consistent with the framework adopted in this research. Furthermore, when the teamwork of primary agents is a reliable means, the legal systems remains on the background as enforcement mechanism. When there is the risk that entrepreneurs do not share valuable information, the legal system steps in, imposing consumer empowering information duties. When this approach is considered insufficient by the monitoring agent, the behaviour of primary agent is regulated by means of consumer protection; hence, in the latter case, the legal system directly protects consumers from entrepreneurs. It is interesting to highlight that, in some occasions, the Commission treats the terms “consumer” and “investor” as synonyms. 94 This semantic interchangeability confirms the conceptual connections between the consumer and the investor identified at pp. 6-12. Consumer policy is concerned not only with the microeconomic market rationale, but also with the macro rationale. In fact, in several occasions the Commission has remarked that consumers’ choices play a key role in fostering innovation and the growth of a sustainable economy. 95 The Commission in 2005 has even observed that also health protection is important as health contributes to productivity and growth.96 b) In corporate governance policy, the property conception of the company finds explicit grounds in EU documents. For example, in Recital (3) of the Shareholders Rights Directive, it is observed that shareholders have paid for their voting rights and should be therefore put in the condition of exercising them. Even stronger is the position taken by the Commission in the 2003 Action Plan: “Shareholders own companies, not management – yet too frequently their rights have been trampled on by shoddy, greedy and occasionally fraudulent corporate behaviour. A new sense of proportion and fairness in necessary”.97 With regard to the models of teamwork, the preferred one is currently based primarily on shareholders empowerment. Empowerment relates mainly to ensuring an effective and informed exercise of voting rights.98 However, in the draft of the revision of the Shareholders Rights Directive, it is possible to see a move towards an enhanced level of protection with regard to directors’ remuneration and related parties transaction.99 In the Takeover Bids Directive, the role of protection is even stronger. Takeovers are a very peculiar moment in the life of a company. It is therefore interesting to look at its regulation in detail with the lenses of the principal multi agent framework. First of all, managers are looked at with suspicion. This makes sense, as it is likely that a successful takeover implies a restructuring of the management. Art. 9 limits greatly their decision making power without the authorization of shareholders to (i) only normal course of business operations and (ii) even these operations only as long as they do 92 European Commission (2005: 3; 2007: 6). 93 OECD (2010: 34-35). 94 European Commission (2002: C 137/8; 2012: 4). 95 European Commission (2007: 2, 3; 2012: 5, 16). 96 European Commission (2005: 2). 97 European Commission (2003: 7). 98 See Directive 2007/36/EC Of The European Parliament And Of The Council of 11 July 2007 on the exercise of certain rights of shareholders in listed companies [2007] OJ L184/18 (Shareholders Rights Directive). 99 See draft articles 9a, 9b and 9c. European University Institute 21 Stefan Grundmann and Fabrizio Esposito not frustrate the bid. Nonetheless, they can search alternative bidders on the market. They also have a duty to evaluate the offer for the shareholders, considering also the impact on occupation (Art. 10). Here we clearly see a case in which non-allocative concerns are taken into account by the legal system. Second, in order to ensure an informed decision by shareholders, information duties and the bidding offer are intensively regulated. The directive clearly shows a concern for the protection of shareholders. In particular, shareholders become agents of each other. More specifically, we see an enhanced version of the competition for active consumers type of teamwork: the highest price paid to one shareholder prior to the bid (no less than six months and no more of twelve months) is the price applicable to all the purchases of shares. This approach implies that the shareholder that has sold his shares at the higher price before the bid ensures that the bidder has the duty to offer the same economic condition to each shareholder. However, the mechanism is not inflexible and the legal system can intervene to modify this price under pre-determined conditions. For the purposes of our argument, it is of interest to note that in corporate governance documents, in some occasions the concept of shareholder is used interchangeably with that of investor.100 Corporate governance policy is clearly not only about shareholders’ interests. As seen, there is a concern for the interest of employees and, more generally, there is a concern for the systemic effects of corporate governance. The 2011 Green Paper shows, for example, a concern for the detrimental effects which investment strategies focusing too often on short-term investments have or can have on growth (“shortermism”).101 The revision of the Shareholders Rights Directive tries to address this issue.102 In conclusion, we have seen that the principal multi agent framework performs quite well in reconstructing the interactions (teamwork) of the legal system (the monitoring agent) with consumers/shareholders and entrepreneurs managers (the two primary agents). The qualification of consumers/shareholders as principal has also explicit grounds. First, consumer policy quite clearly associates consumer empowerment to the autonomy conception of consumer sovereignty and consumer protection to the welfare conception of that concept. As seen, this approach is easily accounted for in a multi agent framework aiming at the conception of allocative efficiency introduced at p. 4. Second, the property conception of the corporation – which, as seen at pp. 11-12, is the normative foundation of the principal conception of shareholders and investors more generally – is explicitly referred to in the EU legal discourse. We have also seen that consumers and shareholders interests are not all that matters. However, we have never claimed that they are. Finally, in both EU policies we have seen a tendency to conflate, in turn, the concept of consumer and that of shareholder, with the concept of investor. It is on these bases that we now turn to discuss in more detail the content of MiFID II, CCD, and MCD. MiFID II and the client relationship as fiduciary relationship: the new disclosure and know your customer regime The models of explanation of fiduciary relationship We start out with the discussion of MiFID II as our first example. There are three reasons for doing so and one requires splitting up this first example into two sections: the first reason is that MiFID II is the most recent reform piece in the direct client relationship in the whole wave of reforms of financial services contract law in the wake of the financial crisis.103 Thus, the most recent trends are likely to be particularly visible here. The second reason is more substantive: MiFID II – as already its predecessors, 100 For example, European Commission (2003: 3) and Shareholders Rights Directive, Recitals (4) and (11). 101 European Commission (2011: 12-13). 102 For an early comment, see Chiu (2016). 103 Directive 2014/65/EC of the European Parliament and of the Council of 15 May 2014 on Markets in Financial Instruments and amending Directives 2002/92/EC and 2011/11/EU, EU OJ 2014 L 173/349. A parallel regulation has been issued for the more technical questions (not really important for the following), also at level I: Regulation (EU) Nr. 600/2014 of the European Parliament and of the Council of 15 May 2014 on Markets in Financial Instruments and amending Regulation (EU) Nr. 648/2012, EU OJ 2014 L 173/84. 22 Department of Law Working Papers Investor-Consumer or Overall Welfare the Investment Services Directive of 1993 and MiFID I of 2004 – is probably the one directive where the view is largely prevalent and perhaps even unanimous – at least on the legal scholarship side – that the relationship is a fiduciary relationship or – tantamount – a principal agent relationship. Indeed, according to Art. 24(1), the general aim of “investor protection” is to ensure that financial entrepreneurs “act honestly, fairly and professionally in the best interests of its clients”. In the following, this will be explained also on the basis of the economic principal agent theory in the founding pieces. The third reason is that this regime is particularly complex and goes as far as not only stating the basic rule and then specifying details of it – the agent has to take the interest of the principal as sole guide-line for her actions –, but that the mere existence of this prime substantive law rule was not seen to suffice. Rather, a second level set of rules was added which is designed to render compliance more likely – i.e. a set of rules which is not about the allocation of rights and duties between the two parties as such, but which is preventive in character (see more in detail below, p. 29). In this sub-section, we will explain the foundations of the substantive rule as such – about the allocation of rights and duties between the parties –, namely the theories advanced to explain the rule that the agent has to take the interest of the principal as the sole guide-line for action. Before doing so, it should be specified that all three major sets of theories consist of an amalgam of theoretical approaches taken from both the legal value hierarchy and from economic theory and that it is often difficult really to segregate one part from the other (if at all, such segregation is helpful or even desirable). The whole set of theories is about justifying the unanimously accepted rule that the agent has to take the interest of the principal as sole guide-line for her actions (see Art. 24(1) MiFID II). We will not focus so much on an in depth analysis of the recitals of MiFID II – even though typically recitals of EU Directives can tell a lot about the reasons of regulation. In this case, however, the recitals are too crude and contain too much only wholesale arguments for the inquiry of a solid theoretical background.104 What the recitals, however, show is that besides fostering the ‘best interest of investors’, MiFID II tries to protect also financial stability and the access to capital of small and medium enterprises. However, the emphasis on the interest of protection of investors is overwhelming in the recitals. The first explanation of the basic fiduciary rule named is about ‘fairness’. The argument runs like this: the client (principal) entrusts a position of influence or information to the investment service provider (agent), and it can therefore be assumed that this entrusted position is handed over with the proviso that it be used only for the benefit of the trustor, i.e. the client.105 This version of the argument has later been refined: the argument that the principal’s interest be the sole guide-line is really convincing – as an implied term of the agreement – (only) in all those cases where the fiduciary receives such position without giving any consideration in exchange. In other words, it can be assumed that the service provider (fiduciary) has to use the position received solely for the benefit of the trustor if the service provider is compensated for his services by a fee (for the rendering of the service) and does not incur any additional cost. In other words: the position of influence or information has not only been entrusted by the client, but has been entrusted ‘for free’ (‘without any compensation’), i.e. in such a way that the service provider does not have other costs for its use than the service as such (which is paid for by fees).106 Given that, in such a situation, the service provider has no vested interest (not to be burdened) at all, the implied term cannot really be doubted that the use of the position entrusted should be solely for the benefit of the trustor. The second explanation of the basic fiduciary rule named is about allocative efficiency, namely about ‘undetectable hidden gains’. The argument runs basically like this: while the client can observe the 104 For an analysis of the recitals, see Grundmann and Hacker (forthcoming, 2016: sub A. I. 2). For a more extended description of the approaches named in the following cf. also this article sub A. II. 105 Most explicitly and with lengthy explanation (and potentially also the first in time): Zöllner (1963: 341-356 et passim); the argument can, however, also be found in the common law literature and case law on fiduciary duties, see from the early development of this idea namely: Shepherd (1981: 35-42, 93-123) (“Encumbered Powers”); critical in this respect, however: DeMott (1988: 912 et seqq.); see as well, with ample reference to case law, Finn (1977); Vinter (1955); Frankel (1983). 106 First so clearly for this additional criterion, to our knowledge, Grundmann (1997: chapters 4 and 5); and summary in English in: Grundmann (1999). European University Institute 23 Stefan Grundmann and Fabrizio Esposito amount of fees, and to some extent also the outcome of the advice given and perhaps even the correctness of it, albeit only in tendency, she can typically not observe at all neither whether another advice would have been better (and would have been given had the service provider not been under the influence of a conflicting interest) nor whether the service provider has profited from the position of influence (to the detriment of the client). Such hidden gains which the service provider makes in addition to the fee openly agreed upon between client and service provider thus render it impossible for the client to assess the overall quality of the service provided, including all its “side-effects” – as compared to other services offered – and the price finally to be paid for it.107 Unless all service providers are forced by regulation (and sanctions) to disclose any other gains than those agreed upon with the fees, neither the quality of service nor the price can be observed – with the consequences first described by G. Akerlof (adverse selection, market for lemons).108 It is just not transparent how high the chances are that the client would have been offered a better investment opportunity without such conflicts of interest existing. In terms of information theory, the problem of vagueness is that the quality of the advice (service) is a credence rather than an experience or even an inspection good.109 This would seem to be the line of arguments which also owes most to principal agent theory as developed by M. Jensen and W. Meckling: 110 the basis of principal agent theory as applied to questions of structure of finance for enterprises would seem to be that the professional side has to make the offers as competitive as possible – to attract the investors –, but as well as transparent as possible to render the choice by the investors as informed as possible (with a view to foster allocative efficiency). From a consumer sovereignty perspective, this is mandated by the guideline that it is the side of consumers which is entitled to the residual gains. In other words, the basic principle in the law on investment services stands in a particularly clear way for a paradigm of consumer sovereignty. The third explanation of the basic fiduciary rule named is about ‘cognitive limitations’. Much of the rhetoric of the recitals of MiFID II points indeed into this direction: this explanation is both about cognitive capacity limits and about bounded rationality in the strict sense. Cognitive capacity limits are about how much information investors can retrieve – of course varying from investor to investor, but relevant if private investment is to be attracted at all – and where, on the other side, information overload starts.111 Beyond that limit, additional information rather deteriorates the quality of decisions.112 Bounded rationality is about cognitive biases – even if the transparency and amount of information are well calibrated. Important biases, for instance with respect to conflicts of interest, may be that the client believes that this service provider would not give in to a conflicting interest (over-optimism).113 By hindsight – and absent massive losses – the client may also filter only or mainly information which ‘confirms’ that she was served by a particularly loyal service provider (confirmation bias).114 One figure proves how important these considerations really are: more than 80 % of the advisees in the EU generally trust their advisors.115 The implication of cognitive limitations is that consumers are often not as capable of fostering their own interests on the (financial) market as we believed them to be. Thus, the legal 107 See MiFID II, Recital (72). 108 Akerlof (1970). For the efficiency implication of hidden gains with respect to the cost of investment advice, see more in detail Bryant and Taylor (2012); Council of Economic Advisers of the President of the United States (2015); Blake (2014); and with respect to the value of the investment advice to the consumer see: Finke (2013: 229). 109 On this (rough) categorization of situations into three paradigmatic cases – in which quality can properly be assessed either before the transaction, or at least by hindsight or not at all –, see more in detail Creane and Jeitschko (2012); as for search goods and experience and credence goods see the pioneering works of Nelson (1970); Darby and Karni (1973); and today Magat and Viscusi (1992); Magat (1998); Furubotn and Richter (2005: 318-326). 110 Jensen and Meckling (1976). 111 See, e.g., Eppler and Mengis (2004). 112 Edmunds and Morris (2000). 113 On optimism bias, see Weinstein (1980); Weinstein and Klein (1996); for a cautionary note from a mathematical perspective, however, see Harris and Hahn (2011). 114 See Lord, Lepper and Preston (1984) (calling the effect “biased assimilation of new evidence”); Nickerson (1998). 115 Chater, Huck and Inderst (2010: 385). 24 Department of Law Working Papers Investor-Consumer or Overall Welfare system has to intervene on new grounds (“behavioural market failures”) in the regulation of financial markets. Ultimately, cognitive limitations fall within the EU concept of vulnerability116 and, more generally, of the consumer as weaker party. All three approaches are not to be seen in isolation from each other. They prove convincing – sometimes jointly, sometimes as diverging modes of explanation – in different situations. The main purpose of the following is twofold: (i) to see how the different approaches prove helpful in different situations, often one or the other being particularly convincing (and also more convincing than the others); and (ii) to see that there is, however, one main thrust which is to use these approaches in general always with one ultimate goal only, which is to foster the interest of the investor(s), and that the main problem with this goal is not the clash with the interests of the service provider, but the fact that not all investors have the same interests (heterogeneity of the group of investors): they have different limits of information retrieval capacity, their cognitive biases are different, but also their capacity to bear losses diverges substantially. Thus, the concept of consumer sovereignty does not face mainly the problem whether it is warranted at all (prevalence over the interests of service providers), but rather the problem of which consumer interests – out of the heterogeneous set of consumer interests – is to be preferred in which situation. It is of great interest in the context of warrantying the framework proposed by this article to compare our understanding of allocative efficiency with the three rationales just presented. The fairness argument connects the fiduciary duty to the idea of delegation of rights. Basically, this is the same normative ground we saw at pp. 11-12 as an alternative to an aggregate value (micro)economic market rationale. As seen, it works both for consumers and shareholders. It is therefore unsurprising that it works for investors too. The argument about hidden gains points out that when gains are hidden, consumers are not in the position to choose in their best interest. Notably, in an overall value framework, comparisons between losses and gains should be either made or their irrelevance should be justified on overall value grounds (for example, by arguing that surplus distribution is irrelevant). However, the hidden costs argument puts the entire focus on the effect of hidden costs on consumers. Indeed, this is the focus of our understanding of allocative efficiency. Moreover, it is the focus of corporate governance – a connection fully consistent with our approach.117 Finally, due to cognitive limitations, informed consumers are not necessarily the best judges of their interests. The implication is that considering cognitive limitations increases the potential scope for market behaviour regulation. This is important because the aggregate value implications of cognitive limitations are rather indeterminate. To the contrary, their implications for a framework aiming at our understanding of allocative efficiency are clearer: the teamwork of primary agents is a less reliable means to the end of allocative efficiency.118 We have seen that the three main rationales for the introduction of fiduciary duties in financial services regulations are all compatible with the notion of allocative efficiency proposed by this framework. We now turn to the analysis of the regulatory choices made in the MiFID II to make our case stronger. Disclosure regime, know your customer and best execution rules as significant examples The basic architecture of the regime – in all its remarkable detail – is threefold (see arts. 24 and 25 MiFID II). It consists of: (i) a disclosure on the offers made – the investment instruments and namely its issuers –, (ii) a clarification of the appropriateness of such instrument for the particular investor served (“know your customer”), and (iii) – when it comes to execution – a duty to choose the “best [way of] execution” possible. For the sake of clarity, we analyse only the basic regime, and despite their high importance not specific roles and situations such as independent investment advice (Art. 24(7) MiFID II), portfolio management (Art. 24(8) MiFID II), packaged investment or insurance-based investment 116 On this connection see, with ample references, Grundmann (2016). 117 See above, pp. 7-14. 118 Elsewhere, one of us has called this implication the “dismality thesis of behavioural sciences” for consumer policy; see Esposito (2017). European University Institute 25 Stefan Grundmann and Fabrizio Esposito products (PRIIPS) (Art. 24(11) MiFID II), and inducements (Art. 24(9) MiFID II etc. (the latter are part of the regime on conflict of interests anyhow and therefore will shortly be taken up below, pp. 29-32). We do not analyse these situations in detail, but just stress that similar considerations to those found in the following could be made for them as well. a) The disclosure on the product itself – namely on the issuer and on the risks of the particular kind of instrument (namely bonds, shares, futures, options and other derivative contracts) – is regulated in Art. 24 MiFID, namely paragraphs 2, 4 and 5. The starting point – remarkable enough – is regulated in a rather hidden way: “Adequate information on the financial instruments” has to be given. In other words: the instruments have to be explained in their functioning, the economic solidity, the chances and the risks they carry. This is remarkable insofar as empowerment of the investor is carried so far that she should be basically put in the position of taking her decision on her best interest on the basis of the relevant (“adequate”) information. Investor sovereignty – in the form of informed investor autonomy – reigns supreme. This rather hidden starting point is remarkable insofar as the burden of gathering the information is not put on the investor at all – contrary to what happens in other markets, at least to a certain extent. Rather the provider of the investment services and its potential as an information intermediary – namely the economies of scale, his typically higher expertise – are abundantly used and profited of. This already rather far-reaching basic rule does, however, not exhaust the regime. Additional rules foster best possible compliance with the basic standard: namely the rules in Art. 24(2) MiFID II. This rule invites providers of investment services to inquire systematically into the properties of such products, to categorize them according to risks, chances and appropriateness for different kinds of investors. In other words, this rule, introduced as a novelty in MiFID II, forces the provider to become systematically aware of the characteristics of the investment instrument. Moreover, in situations where different groups of investors have different interests (see above, p. 27) the MiFID II requires tailor-made solutions: each provider must be clear about which market of investors she is serving with which products (which implies well formulated strategies and careful and thoroughly opined categorization of each and every product). Not only on the side of the product, utmost clarity is required, but also on the side of costs and fees (Art. 24(4) MiFID II) – combatting possibilities of hidden costs from scratch. The whole information regime, moreover, has to be as easily understandable and clear as possible (Art. 24(5) MiFID II). Altogether, all three explanations given for the fiduciary duty can be well traced in this regime: the duty to use the position of trust which the client places in the provider of investment services exclusively in the interest of the client; the duty not to make hidden gains; and as well the philosophy to reduce the effect of cognitive problems of investors best possible. b) Of similar importance is the duty of the provider of investment services to “know your/her customer” (namely Art. 25 MiFID II). This duty is particularly worth mentioning – as it seems unique in the area of investment services, not to be found in other consumer contracts. The provider of investment services has to inquire into the cognitive capacity of each investor needed to do the type of investment envisaged – which is seen to require an inquiry both into the knowledge and into the (practical) experience of each investor –, and moreover into the capacity to absorb risk (financial situation) and the degree to which he is willing to take risks. Again, these items do not only have to be inquired, information has to be given, but all this has to be put down in a protocol, and an overall assessment is owed in the form whether, given the particular characteristics named of each individual investor, the investment is appropriate for this particular investor (see Art. 25(2) and (3) MiFID II). Only for certain groups of (highly sophisticated) investors and (lower risk) transactions, an exception is possible in the form of a transaction “execution only” (Art. 25(4) MiFID). The strict “know your customer” rule – with a clear procedure of the steps to be taken – is particular for the area of investment services, but can be explained also in a broader perspective: in most areas of consumer contract law, the consumer himself is trusted to assess adequately which offer is best for her (once the characteristics of the offer are disclosed to her). In investment services, however – because of their complexity, but potentially also because of the potentially disastrous effect of biases on the economic status of the investor in this area –, more is 26 Department of Law Working Papers Investor-Consumer or Overall Welfare required to foster consumer welfare. It is clear that also this rule – and this rule in particular – is aimed at consumer welfare primarily. c) Once the phase of execution is entered, no longer the investor decides on the measures to be taken, but the provider of investment services. Hereby, the choice of venue, of time, of order in which clients’ orders are executed is all with the provider. Therefore, a particularly refined rule on “best execution” is required – increasingly refined over the years and in the three generations of the EU legal measures (ISD, MiFID I and MiFID II). This regime does not need to be discussed in detail here. Even without such discussion, it is clear that, again, consumer interests reign supreme. MiFID II and its new conflicts of interest regime The relationship to the basic fiduciary model MiFID II – as was already the case of the Investment Services Directive of 1993 and of MiFID I of 2004 – does not stop at stating the basic rule that the investment services provider has to take the interest of the investor(s) as the sole guide-line of her actions. Conversely, it formulates also a nuanced and extensive regime of preventive rules – avoiding as far as possible conflicts of interest, which might endanger compliance with the basic rule named above, or at least neutralizing the effects of conflicts of interest as much possible. And this regime is perhaps even more striking with respect to the topic of consumer sovereignty discussed here. Thus, the regime is peculiar in installing additional compliance mechanisms. In the following, the three most important or most paradigmatic ones are taken up in turn. They are: organizational duties; the disclosure of remaining conflicts of interest; the regime on fees (inducements). The first one is paradigmatic and far-reaching because it does not only impose on investment firms duties with respect to the client, but prescribes as well how to arrange the internal organization with a view to reach this goal in the relationship with the client. This is a rather unique and far-reaching measure. To put it differently, investment services was at least the first area where such regulatory inroad into the board’s competence to shape the organization of the company autonomously became strikingly apparent. The second instrument is paradigmatic because it applies the regulatory approach which is certainly dominant in financial services and was discussed in the last section mainly – the information model – to the particular case of conflicts of interest. The third mechanism is particularly important: it has led to the bulk of case law in Germany, the EU’s strongest economy, and at the same time has triggered the most thoroughly discussed ‘gold plating’ practice in the UK, the EU’s largest financial centre, which in most cases exercises a particularly high influence namely on regulation in financial services. First example: far-reaching organizational duties MiFID II – as was already the case of the Investment Services Directive of 1993 and of MiFID I of 2004 – imposes a duty to take “all reasonable steps designed to prevent conflicts of interest … from adversely affecting the interest of its clients” (Art. 16(3) MiFID II – but note that Art. 16 establishes many other very specific organizational requirements). The main interpretative problem is specifying when “all reasonable measures” have been taken. Following Art. 24(1) and the reference to “the interest of clients” in Art. 16(3) – and consistently with our understanding of allocative efficiency – this duty stops where more organizational arrangements would be more costly than the benefit for “the interests of its clients” of such additional measures. Thus the starting point of the regime on conflict of interests is a company structure solution – because disclosure is seen not to be efficient enough: avoidance of conflicts of interest starts earlier and in a strikingly regulatory way. Under general company law, these aspects would rather be subject of duties of proper organization imposed on the board, but whose arrangement as such and certainly whose details would be left to the board’s discretion.119 Thus, Chinese walls – segregating 119 In relation to the exploration of the legal control on managerial discretion, see Parkinson (1995: 73-96); for an economic analysis of managerial discretion, see Williamson (1963); for an up-to-date empirical oriented paper pertaining to the role European University Institute 27 Stefan Grundmann and Fabrizio Esposito certain departments of banks, namely trading and credit departments from those giving investment advice – is a mandatory requirement and was already under MiFID I. On the one hand, beyond some core prerequisites prescribed specifically, there is only the scope of all endeavors fixed – to minimize the agency costs caused by conflicts of interest via organizational measures – while on the other hand, banks are invited to start their own discovery of the best organization structure.120 Despite the flexibility given, the main thrust is clear: since conflicts of interest raise the mentioned problems of fairness, efficiency, and cognitive limitations on the one hand, and since the mere duty to act in the client’s best interest does not seem sufficient to exhort providers to do so in fact, financial services regulation makes specific provisions concerning the governance and organizational structure of the company. These provisions erect the first line of defense. The overall scope is client welfare – and the client envisaged is primarily the one (group) who could not be sufficiently warned by mere disclosure rules nor ask the right questions to sufficiently assess the risk of conflicts of interest and act accordingly. The symmetry with the structure of consumer and shareholder policies seen at pp. 21-24 is self-evident. Second example: The duty to disclose remaining conflicts of interest Whenever organizational schemes do not reach as far as eliminating the source of conflicts of interest, MiFID II imposes a duty at least to disclose the conflict of interest (Art. 23(2) MiFID II) – as was already the case of the Investment Services Directive of 1993 and of MiFID I of 2004. In this way, while the preventive (and to some extent regulatory) scheme of organizational safeguards comes first, it is not pushed beyond limits of the reasonable121 and the remainder of risk of conflicts of interest is entrusted to the typical information model – based primarily on an assumption of rational information retrieval. Even the information model, however, can be enriched by elements based on insight from behavioural sciences and economics. As has been said, more than half of the investors seem to believe that investment advice is typically given on a fully independent basis. Besides information and warning of the client – to be aware of potential conflicts of interest and their potential impact –, the scheme is also aimed at disciplining providers so that commissions and other sources of conflicts of interest do not lead to biased advice but rather incentivize information seeking and selection of efficient investment instruments.122 This is the background of the new design of the rule: different from what its predecessors ruled, it describes this duty very specifically. As a starting point, Art. 23(2) MiFID II requires that “the general nature and/or sources of conflicts of interest” must be disclosed (just as Art. 18(2) MiFID I already ruled). Different from its predecessors, however, Art. 23(3)(b) MiFID II adds some details and requires that the disclosure shall “include sufficient detail, taking into account the nature of the client”.123 Indeed, the main question seems to come down to whether the “general nature of the conflict of interest” is to be portrayed in more abstract or in more concrete terms. Indeed, disclosure needs not only to clarify, first, that conflicts of interest play an important role, but also, second, which specific risks of biased advice they create. Under MiFID I, the question most disputed was whether the concrete amount of fees of board’s discretion in company’s internal organization see, Wangrow, Schepker and Barker (2015). For the details of regime imposed with respect to organizational prerequisites, see Grundmann and Hacker (forthcoming 2016: sub B. I). 120 The open-endedness of the regime is aimed at providing incentives for innovation and at taping the efficiency of decentralized decision making. Cf. CESR’s Technical Advice on Possible Implementing Measures of the Directive 2004/39/EC on Markets in Financial Instruments, 1st Set of Mandates’, January 2005, at 41. 121 The majority view is that not every possible prudential measure needs to be taken, rather it is deemed sufficient to limit the efforts to those strategies that can reasonably be expected to be undertaken given its size and structure See, e.g., Grundmann (2015c: n. VI 315); contra Kumpan and Hellgardt (2006: 1715-1716). 122 Inderst and Ottaviani (2012). 123 Again, the same provision can be found in the MiFID I ID, Art. 22(4), which was moved up to Level 1 in MiFID II. Therefore, already under MiFID I, a minority view held that disclosure has to be such that the mechanism how the conflict of interest works and why lack of neutrality has to be feared has to be disclosed: see namely, even before introduction of Level 2 legislation: Grundmann (2015c: n. VI 227 already in the editions before and with further references). 28 Department of Law Working Papers Investor-Consumer or Overall Welfare the investment firm gained on one advice and on an alternative one had to be disclosed.124 As the detail was added and as specifying the concrete functioning of the conflict of interest adds both awareness on the side of the provider and on the side of the client (the warning is more consistent), the most convincing interpretation would seem to be that any detail which the concrete client can reasonably ‘digest’ and from which she can profit in her assessment is to be handed over. Namely a warning on any major difference in fees gained should be understandable to all clients. Thus, under the consumer sovereignty paradigm, they would be empowered to decide on whether they judge the risk to be so high that it might have tainted the advice or whether they find it convincing nevertheless. Again, the interest of the clients is the sole guide-line of action. Finally, the new and more specific wording can also be related to a duty to take into account the biases of typical groups of clients (‘nature of client’) and take measures of debiasing. Again, the whole design is clearly focused on furthering the interest of the investor(s). And, again, the main problem with this goal is not the clash with the interests of the service provider, but the fact that not all investors have the same interests (heterogeneity of the group of investors): they have different information needs and different cognitive biases (see more in detail above, pp. 26-27). Third example: The regime on fees (inducements) As has been said already, the regime on fees (the so-called inducements) is of particular practical importance: tt has lead to the bulk of case law in Germany and, at the same time, it has triggered the most thoroughly discussed ‘gold plating’ practice in the UK with its particularly high influence mainly on regulation in financial services – now in MiFID II also with respect to inducements. The cases of the UK (Retail Distribution Review banning commissions) and of Germany (§ 31d Wertpapierhandelsgesetz [WpHG]) are to the point.125 Indeed, inducements is one of the areas where MiFID II reaches well beyond all predecessors. While the traditional scheme contained only one approach, the new regime adds a second one. The problem of inducements and their obvious influence on conflicts of interest, is thus now tackled with a market structure approach: leaving the choice of the regime to the parties, one regime being such that investment firms still can earn on income from sources such as making profits on products conceived by the investment firm itself or subsidiaries, certain kickback fees handed over to them by their transaction partners in execution (but not all!) etc., the other regime, however, aiming at complete neutrality of the intermediary, excluding all typical conflicts of interest and having the client pay in a completely transparent way all the intermediary service. The details of the regime – even though highly interesting from a financial services and functions of intermediaries perspective – would go beyond the scope of this article.126 What is paradigmatic though is the split into two alternatives. It has been substantially criticized because paying very substantial fees for the – then neutral! – service is often not what private investors choose, and therefore markets of independent investment advisors did not or not sufficiently develop for these groups of investors. The assessment of the introduction of a new regime should, however, also focus on considerations following from what has been explained: while the regime in its entirety is aimed at fostering investors’ interests, some parts are primarily aimed at fostering the interests of one group, others the interests of another group primarily. Therefore, the overall regime has to be assessed in its entirety for the most important groups of investors (with interests diverging from each other). Moreover, adding a layer of protection which theoretically has substantial advantages (more neutral advice), but 124 Even though a similar provision as now in MiFID II could be found in the MiFID I 2 nd level Implementing Directives, Art. 22(4), which was moved up to Level 1 in MiFID II, only a minority view under MiFID I held that disclosure has to be such that the mechanism how the conflict of interest works and why lack of neutrality has to be feared has to be disclosed: see namely, also already before introduction of Level 2 legislation and based on the general principle of fiduciary duties: Grundmann (2015, n. VI 227 already in the editions before and with further references). 125 See Papaconstantinou (forthcoming, 2016); for the UK, see also https://www.fca.org.uk/news/post-implementation-reviewof-the-rdr. For Germany, see short survey, also on the large number of decisions rendered by the Supreme Court on Inducements: Grundmann (2015c: n. VI 196, 197 and 288-291). 126 For a detailed analysis (including economic theory and policy assessment of the different parts of the new regime), see Papaconstantinou (last footnote) and Grundmann and Hacker (forthcoming, 2016: sub B. III). European University Institute 29 Stefan Grundmann and Fabrizio Esposito leaving it to parties to opt for this alternative model – namely those parties who typically are very sophisticated – would scarcely seem to be an erroneous strategy (even if success is not complete, but only partial). Therefore one can summarize that the market structure approach chosen for the question of inducements leaves intact a decently functioning (old) regime, but adds as well an approach which, for some groups of investors and under certain conditions, would seem to have substantial theoretical advantages and leaves it to the market (in this case a rather well informed market with sophisticated players) to opt for the second solution where appropriate. Thus the new regime would seem to install a discovery mechanism on the two alternatives (with a good deal of stickiness of the old scheme). Conflicts of interest regulation within the principal multi agent framework From the perspective of the principal multi agent model built above (pp. 14-21), these three tools for the regulation of conflicts of interest are very interesting. Organization duties, as seen above, show that the monitoring agent has taken some decisional power away from financial entrepreneurs and exercised it directly. However, the reallocation has not been complete: to some extent, financial entrepreneurial remain in charge of designing the organization structure. This implies that, to some extent, the legal system considers the financial entrepreneur still an efficient consumer agent. The resulting regime is thus a mix of two approaches: on the one hand, the legal system makes the aim (minimizing the agency costs of conflicts of interests) explicit and leaves to financial entrepreneurs the task of identifying the best means to that end – like in a no delegation model; on the other hand, the legal system establishes some more detailed requirements – like in the full delegation model. The residual duty to disclose implies that while empowerment is not sufficient, in combination with other tools, it is not considered useless either by the EU legislator. The fact that the MiFID II is more detailed is relevant for a level of analysis we do not consider here, namely, the allocation of power in a multilevel legal system like the EU.127 What matters the most here is that MiFID II has inverted the normal relation between empowerment and protection in market regulation. In fact, normally, protection is introduced if empowerment is not considered sufficient. In the regulation of conflicts of interests, instead, protection is the basic strategy and when it does not suffice, it is supplemented by disclosure. Understanding the reformed EU consumer credit law Credit regulation has undeniably a macro dimension. Credit allows entrepreneurs and consumers to engage in activities that would otherwise be too costly to them. However, when credit is given too easily, the level of defaults becomes unsustainable for creditors and financial stability is endangered. Prudential regulation typically aims at safeguarding the stability of the credit system by imposing a number of institutional and conduct constraint on credit providers.128 Importantly, financial stability also has a micro dimension. In its micro dimension, consumers can be harmed in several ways by financial instability. First, financial instability can lead, in one way or another, indebted consumers into overindebtedness or, at least, make them experience financial difficulties. For example, a variable rate loan can become difficult to bear due to financial instability, or the consumer can experience a reduction of income – with obvious consequences on his solvency. Second, instability can also cause a credit crunch, for example, due to the difficulty for financial entrepreneurs of predicting the value of the relevant economic variables or, more fundamentally, for the default risk of financial entrepreneurs which implies a more conservative approach to credit services. When this happens, some consumers will be excluded from the credit market, circumstance that reduces their capacity of intertemporal allocation of budget. For others, the credit conditions will be worsened. Due to instability, on the one hand, the financial status of indebted consumers can worsen and, on the other hand, some consumers can remain excluded from the credit market or can access to it at higher costs. In between, there is the externality problem. A 127 The same applies also to the regulation of the creditworthiness assessment in the Consumer Credit Directive and in the Mortgage Credit Directive; see below, 32-37. 128 Köndgen (2011: 41-43). 30 Department of Law Working Papers Investor-Consumer or Overall Welfare cause of excessive lending in the last financial crisis has been the shifting from a model of “originate to hold” to a model of “originate to distribute” in the mortgage market based on the securitization of loans. This shift of business model has reduced the incentives of borrowers to assess the financial creditworthiness of consumers contributing to the development in the bubble in the real estate market.129 Both the building and bursting of the bubble have had external effects. Hence, there was an externality problem which was relevant both at the micro and the macro level. Given this plurality of goals and their interplay, one way to understand to what extent the interest of the consumers is a concern of credit policies is looking at the extent to which the EU regulation of the credit contract cares about consumers interest in the design of the contract and its execution. First, we look at goals of the 1987 Consumer Credit Directive (CCD87), the current Consumer Credit Directive (CCD) and, finally, at the so called (Consumer) Mortgage Directive (MCD). Second, we look more in detail at the provisions on the creditworthiness assessment an inquiry into their relation with the consumer interest. In fact, in order to establish whether a debtor is worthy enough, you have to identify an outcome that you want to avoid. Third, we look at the regulation of credit contracts with the lenses of the principal multi agent model. The aims of credit regulation in the EU In this section, we look at the normative aims of the three mentioned directives. More precisely, we search for the references to macro and micro goals. Here, our approach is supported if macro goals are supplemented by a specific concern for a fair distribution of value and, more specifically, a distribution to the advantage of the consumer. In the CCD87, increasing the volume of the consumer credit market is conceived of as a win-win-winwin-win game, because it benefits “alike, consumers, grantors of credit, manufactures and retailers of goods and providers of services”. However, there is a general distinctive concern for the fact that “terms of credit may be disadvantageous to the consumer” (not for society as a whole). Instead, in the CCD, there is only a general reference to the “emergence of a well-functioning internal market for credit”.130 The theory of market rationale(s) remains completely implicit here. Finally, the MCD opens with a powerful reference to the macro consequences of the financial crisis.131 The concern, however, is again richer: “The problems identified [in the mortgage markets within the Union] have potentially significant macroeconomic spill-over effects, can lead to consumer detriment, act as economic or legal barrier to cross-border activity and create an unlevel playing field between actors”.132 And the directive aims at “a more transparent, efficient and competitive internal market, through consistent, flexible and fair credit agreements to immovable property, while promoting sustainable lending and borrowing and financial inclusion, and hence providing a high level of consumer protection”.133 Recital (24), on tying and bundling practices, is particularly important for our inquiry. In case of tied services – but not in case of bundle – the service provider does not allow to buy single services separately.134 While bundling can “benefit consumers” (not society as a whole), tying is generally banned by Art. 12 because, as the Recital explains, “may induce consumers to enter into credit agreements which are not in their best interest” (again, not of society as a whole). 129 Comparato (2015: 13). This is by no means an entirely novel problem. As early as 2005, Ramsay identified a problem of the credit market in the “exuberant lending” of financial entrepreneurs than cyclically and lead y competitive forces, expose themselves to an over-optimal level of risk; Ramsay (2005: 61). 130 Recital (7). 131 “The financial crisis has shown that irresponsible behaviour by market participants can undermine the foundations of the financial system, leading to a lack of confidence among all parties, in particular consumers, and potentially severe social and economic consequences”; Recital (3). 132 Recital (4). 133 Recital (6). 134 Art. 4(26) and (27). European University Institute 31 Stefan Grundmann and Fabrizio Esposito Importantly for our purposes, the explicit references the credit regulation enshrines – as the theoretical framework discussed here holds – a macro concern for the enlargement of the market credit as a driver of growth, but also a micro concern for fair contracts in the best interest of consumers. Worthy for whom? Three notions of responsible lending The concepts of responsible lending and creditworthiness assessment offer a particularly telling standpoint for looking at the role the concern for the consumer plays in the design of credit regulation. At least three notions of responsible lending can be conceptually distinguished.135 Each of them associates with different contents of the creditworthiness assessment. It is possible to distinguish a prudential assessment, concerned only with the stability dimension; an anti over-indebtessness assessment, limiting the consumer concern to the harshest consequences of a default; and, finally, a suitability assessment, which focuses on the likely reimbursement of the loan by the consumer. The 2013 World Bank report on Responsible Lending is a good starting point for the discussion. Initially, the report links responsible lending to the problem of over-indebtedness. Notably, for the World Bank, among others,136 the concept of over-indebtedness does not only have a financial dimension – the incapacity of meeting one’s financial obligations – but also a psychological dimension – “the stress that over-indebtedness causes”.137 In the more specific discussion of creditworthiness assessment, the report takes a clear position in favour of the prudential assessment: “The ultimate goal of creditworthiness assessment should be the verification that the borrower has sufficient assets or income to pay back the loan. … The creditworthiness assessment is thus a “creditor-focused” test, looking at the probability that the creditor will be repaid in full”.138 Against this pure macro-based definition, in the next page, it is observed that “the repossession of the collateral is not considered a positive result from the point of view of responsible lending”.139 The obvious and unanswered question is “why does repossession matter?”. In fact, if the repossession ensures the full repayment of the creditor, it is hard to see what the problem is from the perspective of a creditor-focus test. The report then observes that “Understanding the limits in reasonable decision-making, some regulators go further” and introduce a suitability test, which “looks at the lending issue from the borrower’s perspective”.140 The suitability test is composed of three elements: “consumer’s best interest, understanding of the product and longterm affordability”. The best interest test aims at advising the consumer on “which product – if any – is the most suitable to fulfil the specific needs of the consumer”. The understanding “test”141 focuses on the need to disclose and explain adequately the information about “product features, benefits and risks”. Finally, “The test of long-term affordability looks at how long-term risks the consumer may face could influence his ability to repay the loan”. The difference with the creditor-focused test is that the longterm affordability test takes into account “the customer’s future” and the “general economic development”.142 One may wonder why the creditor-focused test does not have to take these elements into account. Be this as it may, one is more fundamentally left to wonder what the connection between these two tests and over-indebtedness is. In fact, it is possible to imagine an intermediate test where – 135 Note that usually consumer lawyers assume that the aim of responsible lending is the protection of consumers from overindebtedness. See Domurath (2015: 155); Ferretti (2015: 349); Mak (2015: 411). In our view, creditworthiness assessment goes beyond the protection from over-indebtedness. As we discuss below, this does not mean that over-indebtedness is satisfactorily addressed by the current EU framework. 136 For a review and a detailed discussion, see Ferretti (2015: 350-356). 137 World Bank (2013: 6). 138 World Bank (2013: 34). 139 World Bank (2013: 35). See also page 45, where the loan-to-value ratio is “seen as a risk-management tool for the lender … rather than a responsible lending tool”. 140 World Bank (2013: 40). The same distinction can be found in the report on consumer credit written by Ipsos and London Economics (2013: 61). 141 Actually, more than a test about the consumer, this is a pure rule of conduct for the borrower. 142 World Bank (2013: 42-43). 32 Department of Law Working Papers Investor-Consumer or Overall Welfare rather than searching for the “most suitable” loan or one that does not harm the creditor – the concern is merely that the consumer does not fall into over-indebtedness. From this perspective, repossession may or may not imply over-indebtedness: it depends on the impact of the repossession on the consumer’s overall economic condition. Now that the three categories have been identified with some precision, it is possible to discuss what kind of creditworthiness assessment has been introduced with the CCD and with the MCD. The CCD establishes that the creditor has the duty to assess the creditworthiness of a consumer “on the basis of sufficient information” without giving any guidance neither on what is the aim of the assessment nor the legal consequences of its negative result nor of the lack thereof. As shown by the Kalhan case,143 this can lead to the paradoxical consequence that, after the implementation of the CCD, the unassessed defaulted debtor has to pay the entire loaned sum immediately and with an interest rate substantially identical to the contractual one. This generic provision on responsible lending is unsurprising if one considers the odyssey of this directive (proposed in 2002 and approved with weakening revisions in 2008).144 Be this as it may, at least one element of the suitability test described in the World Bank report is present. It consists in an extensive regulation of information disclosure – which is rather interesting from the perspective of behavioural analysis145 – which includes a duty to explain (Art. 5(6)). The CCD thus contains the World Bank’s report understanding test. Indeed, this is not a great result, as information disclosure is “only” the core of one part of a solid regime, namely of the empowerment pillar of consumer policy. The MCD has a much more structured notion of responsible lending. Moreover, in Recital (55) “Member States … should be encouraged to implement the Financial Stability Board’s Principles for Sound Residential Mortgage Underwriting Practices”. Accordingly, the European Banking Authority has issued in 2015 the Guidelines on creditworthiness assessment (EBA Guidelines). Notably, the EBA Guidelines require “the competent authorities and financial institutions [to] make every effort to comply with the guidelines” on the grounds of Art. 16(3) of Regulation 1093/2010.146 The pressing of EU institution on creditworthiness assessment in the MCD is thus incomparable with the timid reference to it in the CCD. The creditworthiness assessment in the MCD contains two of the three elements of the World Bank’ report suitability test. First, like in the CCD, there is the understanding test: extensive information disclosure duties and also the duty to explain (Art. 16). Second, all creditworthiness assessments must focus on the long-term affordability of the credit, considering “all necessary and relevant factors that could influence a consumer’s ability to pay”.147 As, seen the complete version of the report’s suitability test requires also to assess the credit in the best interest of the consumer. In the MCD, this duty is detached from the creditworthiness assessment, and it constitutes the core feature of advisory services (Art. 22). There are several interesting obligations from a principal multi agent perspective, but what matters here is the definition of advising “in the best interest of the consumer” in Art. 22(3) let. d). It requires to be informed on “the consumer’s needs and circumstances”, give a copy of the recommendation on a durable medium and, more fundamentally, “recommend suitable credit agreements”. The suitable recommendation requirement is satisfied if the advisor recommends “a suitable credit agreement or several suitable credit agreements”. To some extent, this approach is straightforwardly compatible with the framework proposed in this article. There is a teleological relation between the advisor and the consumer’s best interest.148 However, to act in the “best interest” does not 143 Case C-565/12, ECLI:EU:C:2014:190. 144 For a reconstruction, see Rott (2009: 184-187). 145 It is in fact evident that, compared to the Unfair Terms and the Unfair Commercial Practices Directives, the CCD offers solutions more compatible with behavioural insights. 146 European Banking Authority (2015: 7). 147 Recital (55). See European Banking Authority (2015: 10-11). 148 It would require a detailed discussion, but it suffices to note that in the MCD terms like preference and objective on the one hand and need and interest on the other hand, which refer to completely different notions of value, are used interchangeably. European University Institute 33 Stefan Grundmann and Fabrizio Esposito require the suggestion of the most suitable, but simply one or more suitable alternatives. While this seems a contradiction of the normative premise of our discussion, the next section actually show that this is compatible with a certain conception – perhaps not entirely convincing – of the consumer as agent within a principle multi agent model identifying allocative efficiency (as defined here) as the principal. Before moving to this topic, a word is necessary on what, for many, is the core of consumer credit market regulation, namely, fighting over-indebtedness. There is a cursory reference to it in the CCD.149 Overindebtedness is more present in the MCD, although its presence is quite peculiar. It is mentioned in two recitals, in connection with the creditworthiness assessment and the repayment of the outstanding debt after foreclosure proceedings. It is however present in only one article, Article 45, which commits the Commission to present by 21 March 2019 “a comprehensive report assessing the wider challenges of private over-indebtedness linked to credit activity” and, if appropriate, “legislative proposals”. Even if the explicit references are limited, we disagree with the conclusion that EU institutions are patently doing too little to fight over-indebtedness.150 Indeed, we take the point that over-indebtedness is often a problem based on factors external to the credit relation – such as, reduction of income, divorce, taxes, etc.151 This, however, means that credit regulation should be considered in the broader institutional context of a country.152 In our view, it is exactly because of this nature of the problem that the instrumentalization of the credit agreement as a means to fight over-indebtedness cannot be underestimated. First of all, the highly protective standard of creditworthiness assessment cannot be disregarded. In the 2013 Study on the EU credit market, it was observed that in those countries not imposing the duty of assessing credit suitability, the level of consumer default was doubled.153 To this regard, the problem is to design the regulation of the credit market that is allocatively optimal.154 Second, and perhaps more importantly, we cannot ignore the duty imposed on creditors related to the execution of the contract. Not only creditors have the duty to obtain the best price for the foreclosed estate if this price “affects the amount owed by the consumer” (Art. 18(5)). They are also “encourage(d) [to] exercise reasonable forbearance before foreclosure proceedings is initiated” (Art. 18(1)). In light of the above, it can safely be concluded that indeed there is a macro concern in creditworthiness assessment. However, there is also a specific concern for the detrimental consequences for consumers of a default. What remains implicit in the normative framework is the trade-off between the benefits from consumers avoiding a default thanks to the creditworthiness assessment and the costs for consumers that are excluded from the consumer credit market. Importantly for our framework, even when one focuses on this problem the reference to aggregate consumer welfare is clearly much more precise than the reference to overall welfare – as we see in the following. EU regulated credit market as a principal multi agent interaction The aim of this section is showing the analytical potential of the principal multi agent model described at pp. 14-21. In our view, many features of the CCD87, CCD and MCD can be systemically discussed and compared easily within this model. In particular, the model helps to frame some regulatory choices which at first sight are puzzling and to spell out the factual assumptions under which these choices are convincing. As seen, we consider consumers and entrepreneurs as primary agents and the legal system as monitoring agents. The legal system steps in to correct defects in the teamwork of the primary agents. The main strategies of teamwork of primary agents are three: moderately self-interested entrepreneurs, consumer screening entrepreneurs and entrepreneurs competing for active consumers. 149 Informing the consumer about the risks of over-indebtedness is an example of responsible lending (Recital (26)). 150 Domurath (2015). 151 Micklitz (2015: 230-231). 152 Mak (2015) and Kempson (2015). 153 Ipsos and London Economics (2013: 61). 154 Köndgen (2011: 44). 34 Department of Law Working Papers Investor-Consumer or Overall Welfare First of all, EU credit market regulation tries to implement the competition for active consumers type of teamwork by strengthening the Single Market.155 The means to this end are primarily information disclosure duties, that have become more and more elaborate in time.156 These duties are supplemented by a commitment to foster consumer financial education.157 On the supply side, EU law limits the entry to the market of entrepreneurial agents that have a sufficient expertise. This approach is already an intrusive form of monitoring. Perhaps less intrusive, but more recent, is the attention paid to incentive regulation. The MCD tries to avoid that the remuneration mechanism of entrepreneurial agents pushes them to behave in disaccord with the consumer interest.158 Finally, there are the protective duties. They are basically absent in the CCD87. The least intrusive protective duty is perhaps the duty to maintain the offer irrevocable or to grant a right to withdraw in order to give the consumer a “sufficient time to compare offers, assess their implications and make an informed decision”.159 Enhancing competition for active consumers is clearly the end. The second is the advisory service, introduced only by the MCD. While not mandatory, financial entrepreneurs have the duty to inform if they provide it and, if they do, several elements of the transaction are regulated.160 Apart from the regulation of the termination of openend credit agreements in Art. 13(1), the only other form of protection introduced by the CCD is the right to early repayment, granted also by the MCD. This right is more pronounced in the CCD than in the MCD.161 In fact, contrary to the CCD, the MCD introduces several layers of consumer protection.162 For our purposes, two aspects are particularly important: the duty of performing the creditworthiness assessment (which, if negative, imposes the denial of credit163) and to compare how the best interest of consumers is enshrined in the ban of tying practices and in the regulation on advisory services.164 Creditworthiness assessment is viewed by legal scholars as the imposition of a fiduciary duty on financial entrepreneurs which is “atypical … to the commercial contract”.165 Indeed, we see clearly that there is a higher burden on the financial entrepreneur than on the consumer if one compares the duty of responsible lending with the duty of responsible borrowing. While responsible lending requires the creditworthiness assessment, responsible borrowing requires only the duty to disclose information in a non-intentionally incomplete or falsified way.166 We can now move to the second point, comparing how the best interest of consumers is enshrined in the ban (with some exception) of tying practices and how it is enshrined in the regulation on advisory services. In the first case, the approach is the ban of the practice. What financial entrepreneurs have to do is, ultimately, offering two or more financial products both together and separate. This obligation is easy to enforce. Besides, the burden to entrepreneurs is low because they simply have to offer the products together, but also separate. In the case of the advisory 155 See CCD87, Recitals (1)-(4); CCD, Recitals (4), (6) and (7); MCD, Recitals (3)-(6). 156 Compare CCD87, Artt. 3-6 with CCD, Artt. 4-6, 10-12. 18-19, 21 and with MCD, Artt. 10, 11, 13-17, 20, 22-24, 27. 157 CCD, Recital (26); MCD, Art. 6. 158 Art. 7(2)-(4). 159 MCD, Art. 14(6). The CCD grants a right to withdraw of 14 days (Art. 14), while in the MCD the “evaluation period” is of 7 days only. 160 Artt. 15(1) and 22. 161 The default mechanism in Art. 16 CCD is a compensation of either 0,5% or 1% of the early repaid credit, but it can be designed in a way that is even further to the advantage of the consumer. The right can also be tightened. In this case, the creditor can only claim the difference between the original interest rate and the rate available now on the market, plus administrative costs. The amount can never be higher than the original interest. Art. 25 MCD is vaguer. The right can be subject to limitations and the creditor may be “entitled to fair and objective compensation, where justified” that cannot be superior to “the financial loss of the creditor” (note that “financial loss” is not defined in the MCD). 162 See MCD, Artt. 7, 12, 18, 19, 23, 25, 28. 163 Art. 18(5). The MCD does not specify the legal consequences of the breach of this duty. For an interesting comparative analysis of different solutions, see Atayer (2011). 164 See, in particular, Artt. 12 and 22. 165 Ferretti (2015: 362); Köndgen (2011: 49). 166 Artt. 18(4) and 20(1). European University Institute 35 Stefan Grundmann and Fabrizio Esposito service, we have seen that one may wonder why the advisor does not have to suggest “the most suitable” or rank the suitable products in order of convenience. An explanation can be found at the level of institutional capabilities. As seen, financial agents have to comply with specific duties of financial competence. Courts, to the contrary, do not. Therefore, to include “the most suitable” or the ranking criteria in the advising service would have greatly extended the courts’ monitoring powers and conversely put an enormous burden on the service providers. If this is the explanation, the issue then can be analysed in terms of its soundness. Ultimately, the MCD version of the suitability test enshrined in the advisory service leaves it up to the consumer to decide which financial product is the most suitable. In our framework, the implicit normative evaluation is that consumer welfare is going to be higher under the MCD version of the suitability test than under “the most suitable” version of test. This would require consumers to outperform the interplay of the duty on the advisor and the screening of the legal system in terms of allocative efficiency. This article does not aim at offering an assessment of the allocative efficiency (in our sense) of these legal choices. What suffices here is to point out that the framework offers a clear articulation of the different versions of the test. An articulation that could be used for economic analyses of the allocative efficiency of the alternative tests. This section has shown that not only conceptually and potentially, but also normatively and in EU practice, in consumer credit policy the concern goes beyond the macro dimension and the focus is also on (avoiding) the detriment to consumers. In particular, it was seen that the EU notion of credit worthiness assessment clearly goes beyond the interest for the protection of the credit –and, ultimately, of financial stability and growth – to consider the detrimental consequences of default for consumers. Here, the two points of concern remain the implicit trade-off between the costs and benefits for consumers of being excluded from the market in order to avoid default and the extent to which EU law is doing enough to contrast over-indebtedness. Finally, we have seen how the principal multi agent framework introduced at pp. 14-21 proves useful in reconstructing the structure of the regulation of consumer credit markets. Not only the framework explains much with less, but it also helps to identify regulatory choices that do not appear particularly convincing. Conclusions In this article we have offered several arguments aimed at proposing a conception of the microeconomic market rationale (allocative efficiency) that is not concerned with the maximization of overall value, but with the maximization of consumer aggregate welfare through mutual Pareto optimal exchanges. Accordingly, regulated market behaviour can be conceived of as a principal multi agent interaction where allocative efficiency is the principal. This conception is not the result of our own intuition, but it has deep roots in economic literature. The disagreement between the mainstream rational (overall value) and the notion of allocative efficiency we proposed is therefore a disagreement internal to economic theory. Moreover, as our notion of allocative efficiency was embraced, among others, by Adam Smith, it can hardly be considered a disagreement between mainstream thought and heterodox theory. Arguably, the notion of allocative efficiency defended here can be best understood as a sort of ‘hidden mainstream’ microeconomic market rationale. As also discussed, a just society cannot be reduced to a general equilibrium that is allocative efficient. Not only there is a macroeconomic market rationale (growth) to consider, but also other societal values have to be considered. Indeed, much more work is required to offer such a framework. However, this article has shown that the approach proposed here offers a promising explanatory framework of EU market behaviour regulation. In these concluding remarks we point out some of the features of this broader project. First, we understand this project as an example of what Calabresi has recently called Law and Economics and opposed to Economic Analysis of Law.167 While mainstream Economic Analysis of Law is subject 167 36 Calabresi (2016: especially, 1-23). For an insightful legal philosophical analysis of the distinction, see Chiassoni (2016). Department of Law Working Papers Investor-Consumer or Overall Welfare to several critiques,168 Calabresi identifies a key difference between the two approaches relates to the way in which disagreements between the chosen economic theory and the studied legal system are solved. For Economic Analysis of Law, the disagreement is solved invariably in favour of the economic theory. However, ignoring the worldview enshrined in legal concepts is clearly unacceptable and ultimately counterproductive from a legal practitioner. . A Law and Economics approach, to the contrary, is well-suited for interdisciplinary dialogue. Its competitive advantage is that it takes the arguments of both law and economics seriously, more precisely: the arguments of both disciplines are considered as mutually (hermeneutically) influencing each other. Thus, if market behaviour regulation is better understood as a means to maximize the aggregate consumer welfare under the Paretian constraint than to maximize total value, then the lawyer-economist should start building his theory on the former concept rather than on the latter. If market behaviour regulation has two pillars – allocative efficiency and growth – then a theory considering both is even better; and so on. Importantly, to build the argument that allows to choose in a reasoned way between alternative market rationales, it is not sufficient – as it has been done here – to focus only on the explicatory power of one theory. Instead, the explicative power of alternative theories needs to be compared. For example, why does MiFID II clearly refer to the investor’s best interest according to the mainstream economics of law theory? The comparison of theoretical explanations raises complex issues about the interplay of different theoretical desiderata (namely moral plausibility and fitness with the practice) that cannot be discussed here. It can be noted, however, that this issue is not new neither in legal theory in general nor with specific reference to economic explanations of the law.169 Second, the scope of the theory – market behaviour regulation – needs some comments. On the one hand, the focus here is on regulation. Indeed, our primary analytical device, the principal multi agent framework, could be used also for building predictions (and rankings) of alternative allocative performances under different assumptions. Nonetheless, in our view, that is the task of economists. As lawyers, our primary interest is understanding what are the assumptions embodied in legal texts and practices. Thus, rather than on market behaviour, our focus is on the regulation of market behaviour. On the other hand, the identification of the regulated “market” has been left to EU institutions. This approach is in open contrast with the idea – very strong, for example, in the Chicago School – that the market represents an analytical tool applicable to a variety of social contexts normally not considered markets (like sex, marriage and adoptions, to name some of the most striking ones). We do not intend to implicitly reject that approach here. We simply think that before extending a framework about markets beyond the usual understanding of markets, it makes sense to focus on social practices that unambiguously are considered instances of market behaviour. Similarly, in future works we will have to be more precise in the use of terms like consumer, investor, client and shareholder. Finally, this project seems promising also for scholars not necessarily interested in a Law and Economics Theory of (EU) Market Behaviour Regulation, but in the systematic integration of principal agent theory in contract legal scholarship more generally. In fact, one of us has recently called for the creation of the field of legal research on contract governance.170 The main argument was, similarly to the case of Gillette as well as Sunstein and Bar-Gill’s articles,171 that contract governance is a useful – arguably, the best available – framework for building a contract theory allowing to reflect on the most challenging problems of current contractual practices, namely third-parties effects, networks of contracts and longterm contracts. Integrating the contract governance project and the framework discussed here has theoretical and normative advantages. Theoretically, the advantage of the approach proposed in this article is that it allows to analyse also spot contract in their normal and internal dynamic and not only 168 For a survey and the development of new arguments, see Esposito (forthcoming: chapter 1). 169 On the former, see Bix (2013). On the latter, the challenge moved by Dworkin (1980: 219-223) to mainstream theory still has to be met. Arguably, the best attempt was offered by Kraus (2007). For a discussion, see Esposito (forthcoming: chapter 1). 170 See Grundmann, Möslein and Riesenhuber (2015). 171 See above, pp. 20-21. European University Institute 37 Stefan Grundmann and Fabrizio Esposito with respect to the pathology of third-parties effects. Normatively, the framework proposed here also suggests what the normative foundation of the contract governance project could be: allocative efficiency, understood – as suggested here – as maximization of aggregate consumer welfare through mutual Pareto optimal exchanges. While it is hard to predict where this research will lead us to, we are sure that it is going to be a challenging and controversial long-term project. 38 Department of Law Working Papers Investor-Consumer or Overall Welfare References M. D. Adler, Well-Being And Fair Distribution: Beyond Cost-Benefit Analysis, (New York: Oxford University Press, 2012). M. Aglietta and A. Rebérioux, Corporate Governance Adrift: A Critique Of Shareholder Value, (Cheltenham; Northampton, MA: Edward Elgar, 2005). G. Akerlof, ‘The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism’, 84 Quarterly Journal of Economics 488-500 (1970). A. Alchian and H. Demsetz, ‘Production, Information Costs, and Business Organization’, 72 The American Economic Review 777-795 (1972). F. Allen, ‘Stock Markets And Resource Allocation’, in: C. Mayer and X. Vives (eds.), Capital Markets and Financial Intermediation, (New York: Cambridge University Press, 1993), 81108. J. Armour, H. Hansmann and R. Kraakman, ‘What Is Corporate Law’, in: R. Kraakman et al., The Anatomy Of Corporate Law: A Comparative And Functional Approach, (Oxford; New York: Oxford University Press, 2nd edn. 2009) 1-34. K. J. Arrow, ‘Uncertainty And The Welfare Economics Of Medical Care’, 53 American Economic Review 941-973 (1963). Y. Atamer, ‘Duty of Responsible Lending‘, in: S. Grundmann and Y. Atamer (eds.), Financial Services, Financial Crisis, and General European Contract Law – Failure and Challenges of Contracting, (Alphen aan den Rijn: Kluwer, 2011), 179-202. E. Avgouleas, Governance of Global Financial Markets. The Law, The Economics, The Politics, (Cambridge: Cambridge University Press, 2012). O. Bar-Gill and C. R. Sunstein, ‘Regulation As Delegation’, 7 Journal of Legal Analysis, 1-36 (2015). A. Berle and G. Means, The Modern Corporation and Private Property, (New York; Chicago; Washington: Commerce Clearing House Publishing, 1932). B. H. Bix, ‘John Austin and Constructing Theories of Law’, in: M. Freeman and P. Mindus, The Legacy of John Austin’s Jurisprudence, (Dordrecht: Springer, 2013) 1-13. D. Blake, ‘On the Disclosure of the Costs of Investment Management’, Pensions Institute Discussion Paper PI-1407 1-11 (2014). M. Blair and G. Walker, Financial Services Law, (Oxford: Oxford University Press, 2006). C. Bryant and G. Taylor, ‘Fund Management Charges, Investment Costs and Performance’, Statistics Series Paper 3 (Investment Management Association, 2012). P. Bolton and M. Dewatripont, Contract Theory, (Cambridge, Mass.: The MIT Press, 2005). J. M. Buchanan, ‘Economists and the Gains from Trade’, 9 Managerial and Decision Economics, 5-12 (1988). J. Buchanan, D. Heesang Chai, and S. F. Deakin. Hedge Found Activism: the Limits of Shareholder Supremacy (Cambridge: Cambridge University Press, 2012). G. J. Bymers, ‘Consumerism’, in: D. Greenwals (ed.), Encyclopedia of Economics. (New York: McGraw-Hill, 1982). European University Institute 39 Stefan Grundmann and Fabrizio Esposito G. Calabresi, The Future of Law and Economics: Essays in Reform and Recollection (Princeton: Princeton University Press, 2016). C.-W. Canaris, Bankvertragsrecht, (Berlin: de Gruyter, 3rd edn. 1988). N. Chater, S. Huck and R. Inderst, Consumer Decision-Making in Retail Investment Services: A Behavioural Economics Perspective. Final Report, (2010) available at www.vselee.cz/files/useruploads/eu_consumer_behaviour_final_report.pdf. P. Chiassoni, ‘La Terra Promessa del Professor Calabresi, 32 Contratto e Impresa, 608-615 (2016). I. H.-Y. Chiu, ‘European Shareholder Rights Directive Proposals: A Critical Analysis In Mapping With The UK Stewardship Code?’, 17 ERA Forum, 31–44 (2016). The Committee for the Prize in Economic Science in Memory of Alfred Nobel, Oliver Hart and Bengt Holmström: Contract Theory, (2016) available at https://www.nobelprize.org/ nobel_prizes/economic-sciences/laureates/2016/advanced-economicsciences2016.pdf. G. Comparato, ‘The Design of Consumer Mortgage Credit Law in the European System’, in: H.-W. Micklitz and I. Domurath, Consumer Debt and Social Exclusion in Europe, (Farnham, Surrey, England; Burlington, VT, USA: Ashgate, 2015) 9-26. Council of Economic Advisers of the President of the United States, The Effects Of Conflicted Investment Advice On Retirement Savings, (2015) available at https://www.whitehouse .gov/sites/default/files/docs/cea_coi_report_final.pdf. T. Cowen and A. Tabarrok, Modern Principles of Economics, (New York: Worth and Basingstoke, 2011). A. Creane and T. D. Jeitschko, ‘Endogenous Entry in Markets with Unobserved Quality’, Economic Analysis Group Discussion Paper 12-6, 1-37 (2012). K. J. Cseres, Competition Law And Consumer Protection, (The Hague: Kluwer Law International, 2005). M. Darby and E. Karni, ‘Free Competition and the Optimal Amount of Fraud’, 16 Journal of Law and Economics 67-88 (1973). P. Davies et al., Corporate Boards in Law and Practice: A Comparative Analysis in Europe (Oxford: Oxford University Press, 2013); A. De Strobel, La Sovranitá del Consumatore, (Padova: Cedam, 1970). D. DeMott, ‘Beyond Metaphor - an Analysis of Fiduciary Obligation’, 1988 Duke Law Journal 879-924 (1988). D. W. Diamond and R. E. Verrecchia, ‘Optimal Managerial Contracts and Equilibrium Security Prices’, 37 Journal of Finance 275-287 (1982); I. Domurath, ‘A Map of Responsible Lending and Responsible Borrowing in the EU and Suggestions for a Stronger Legal Framework to Prevent Over-Indebtedness of European Consumers’, in: H.W. Micklitz and I. Domurath, Consumer Debt and Social Exclusion in Europe, (Farnham, Surrey, England; Burlington, VT, USA: Ashgate, 2015) 155-175. A. Dorresteijn et al., European Corporate Law, (Alphen Aan Den Rijn: Kluwer Law International, 2009). R. Dworkin, ‘Is Wealth a Value?’, 9 The Journal of Legal Studies, 191-226 (1980). A. Edmunds and A. Morris, ‘The Problem Of Information Overload In Business Organizations: A Review Of Literature’, 20 International Journal of Information and Management, 17-29 (2000). 40 Department of Law Working Papers Investor-Consumer or Overall Welfare M. J. Eppler and J. Mengis, ‘The Concept of Information Overload: A Review of Literature from Organization Science, Accounting, Marketing, MIS, and Related Disciplines’, 20 The Information Society: An International Journal, 1-20 (2004). R. A. Epstein, ‘Harmonization, Heterogeneity And Regulation: CESL, The Lost Opportunity For Constructive Harmonization’, 50 The Common Market Law Review 207-224 (2013). F. Esposito, ‘Alcune Note su di un Approccio Economico Ordinalista allo Studio del Diritto’ 2/2013 Ars Interpretandi 151-184 (2013). F. Esposito, ‘Efficienza Paretiana ed Efficienza Allocativa: Considerazioni Teoriche per Giuesconomisti’, 1/2016 Materiali Per Una Storia della Cultura Giuridica del Novecento 225236 (2016). F. Esposito, ‘A Dismal Reality: Behavioural Analysis and Consumer Policy’, Journal Of Consumer Policy published online (2017). F. Esposito, PhD Thesis (Florence: European University Institute, forthcoming). European Banking Authority, ‘Guidelines On Creditworthiness Assessment. Final Report’, (Guidelines), EBA/GL/2015/11. European Commission, ‘Consumer Policy Action Plan 1999-2001’, (Communication), COM (99). European Commission, ‘Consumer policy strategy 2002-2006’, (Communication), COM (2002) 208 final. European Commission, ‘Modernising Company Law and Enhancing Corporate Governance in the European Union - A Plan to Move Forward’, (Communication), COM (2003) 284 final. European Commission, ‘Healthier, safer, more confident citizens: a Health and Consumer Protection Strategy’, (Communication), COM(2005) 115 final. European Commission, ‘A European Consumer Agenda - Boosting Confidence And Growth’, (Communication), COM(2012) 225 final. European Commission, ‘EU Consumer Policy strategy 2007-2013. Empowering Consumers, Enhancing Their Welfare, Effectively Protecting Them’, (Communication), COM(2007) 99 final. E. Fama and M. Jensen, ‘Separation of Ownership and Control’, 26 Journal of Law and Economics, 301-325 (1983). M. Finke, ‘Financial Advice: Does it make a difference?’, in: O. S. Mitchell and K. Smetters (eds.), The Market for Retirement Financial Advice (Oxford: Oxford University Press, 2013) 229-248. F. Ferretti, ‘The Over-Indebtedness Of European Consumers: Time For A ‘Fresh-Start’ Of The EU Policy And Legal Agenda?’, 11 ERCL 346-375 (2015). P. Finn, Fiduciary Obligations, (Sydney et al.: Law Book Company, 1977). T. Frankel, ‘Fiduciary Law‘, 71 California Law Review 795-836 (1983). E. G. Furubotn and R. Richter, Institutions and Economic Theory. The Contribution of the New Institutional Economics (University of Michigan Press: Ann Arbor, 2nd edn., 2005). F. Gevurtz, Corporation Law, (Eagan: West Group, 2nd edn. 2010). C. Gillette, ‘Rolling Contracts As An Agency Problem’, 2004 Wisconsin Law Review, 679-722 (2004). S. Grundmann, Der Treuhandvertrag – die werbende Treuhand, (Munich: Beck, 1997). European University Institute 41 Stefan Grundmann and Fabrizio Esposito S. Grundmann, ‘Trust and Treuhand at the End of the 20th Century – Key Problems and Shift of Interests’, 47 The American Journal of Comparative Law, 401-428 (1999). S. Grundmann, ‘On the Unity of Private Law – From a Formal to a Substance Based Concept of Private Law’, 2010 ERPL 1055-1078 (2010). S. Grundmann (with the partial contribution of F. Glasow), European Company Law – Organization, Finance and Capital Markets, (Cambridge; Portland, Ore.: Intersentia, 2nd edn. 2012). S. Grundmann, Bankvertragsrecht: vols. 10/1, 10/2 and 11/1 of Großkommentar Handelsgesetzbuch, (Berlin and New York: de Gruyter, 5th edn. 2015, 2016, 2017). S. Grundmann, ‘Akteure in Organisationen – Thema, Theorien und Context, Diskussion’ in: S. Grundmann, H. Micklitz and M. Renner, Privatrechtstheorie, Vol. II, (Tübingen, Mohr-Siebeck, 2015a) 1507-1527, forthcoming also in English. S. Grundmann, ‘Prinzipalsentsscheidung: Abwanderung und Widerspruch – Thema, Theorien und Context, Diskussion’ in: S. Grundmann, H. Micklitz and M. Renner, Privatrechtstheorie, Vol. II, (Tübingen, Mohr-Siebeck, 2015b) 1585-1608, forthcoming also in English. S. Grundmann, ‘Wertpapierhandelsgesetz [Investment Services Act]’, in: Ebenroth, Boujong, Joost and Strohn, Handelsgesetzbuch, (Munich: Beck, 3rd edn. 2015c) n. VI 1-403. S. Grundmann, F. Möslein and K. Riesenhuber (eds.), Contract Governance: Dimensions In Law And Interdisciplinary Research, (Oxford; New York : Oxford University Press, 2015). S. Grundmann and P. Hacker, ‘Conflicts of Interest: Theory and the Regime of MiFID I and II’, in: D. Busch and G. Ferrarini (eds.), MiFID II and the Regulation of European Financial Markets, (Oxford. Oxford University Press, forthcoming 2016). R. Inderst and M. Ottaviani, ‘Regulating Financial Advice’, 13 European Business Organization Law Review 237-246 (2012). H. Hansmann and R. Kraakman, ‘The End Of History For Corporate Law’, 89 Georgetown Law Journal 439-468 (2001). H. Hansmann and R. Kraakman, ‘What Is Corporate Law?’, in: R. Kraakman et al., The Anatomy Of Corporate Law: A Comparative And Functional Approach, (Oxford; New York: Oxford University Press, 2004) 1-31. A. G. L. Harris and U. Hahn, ‘Unrealistic Optimism about Future Life Events: A Cautionary Note’, 118 Psychological Review 135-54 (2011). O. Hart, ‘Corporate Governance: Some Theory and Implications’, 105 The Economic Journal 678-689 (1995). O. Hart, ‘Financial Contracting’, 39 Journal of Economic Literature 1079-1100 (2001). F. von Hayek, ‘Present State of the Debate’, in: F. von Hayek (ed.), Collectivist Economic Planning, (London: Routledge & Kegan Paul Ltd, 1933) 201-243. F. von Hayek, ‘The Meaning of Competition’, in: Caldwell (ed.), The Market and Other Orders, (Chicago: University of Chicago Press, 2014 (1948)) 105-116. J. Heath, Morality, Competition, And The Firm: The Market Failures Approach To Business Ethics, (New York, NY: Oxford University Press, 2014). D. Hildebrand, The Role Of Economic Analysis In EU Competition Law: The European School, (Alphen aan den Rijn, The Netherlands: Kluwer Law International, 4th edn. 2016). 42 Department of Law Working Papers Investor-Consumer or Overall Welfare A. O. Hirschman, Exit, Voice, And Loyalty; Responses To Decline In Firms, Organizations, And States, (Cambridge, Mass.: Harvard University Press, 1970). J. Hirshleifer, Price Theory and Applications, (Englewood Cliffs: Prentice Hall, 1976). B. Holmström, ‘Moral Hazard In Teams’, 13 The Bell Journal of Economics 324-340 (1982). B. Holmström and J. Tirole, ‘Corporate Control And The Monitoring Role Of The Stock Market’, Yale School of Organization and Management Working Paper Series D No. 48 (1990). K. Hopt, ‘Common Principles of Corporate Governance in Europe’, in: B. Markesinis (ed.), The Clifford Chance Millennium Lectures – The Coming Together of the Common Law and the Civil Law, (Oxford: Hart Publisher, 2000), 105-132. W. H. Hutt, Economists and the Public Interest. A Study of Competition and Opinion, (London; Toronto: Jonathan Cape, 1936). W. H. Hutt, ‘The Concept of Consumers’ Sovereignty’, 50 The Economic Journal, 66-77 (1940). Ipsos and London Economics, Study on the Functioning of the Consumer Credit Market in Europe. Final Report, (Brussels: European Commission, 2013). M. Jensen and W. Meckling, ‘Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure’, 3 Journal of Financial Economics 305-360 (1976). C. Jolls, ‘Behavioral Economics Analysis of Redistributive Legal Rules’, 51 Vanderbilt Law Review, 1653-1678 (1998). L. Kaplow, The Theory Of Taxation And Public Economics, (Princeton, N.J.: Princeton University Press, 2008). L. Kaplow, S. Shavell, ‘Why The Legal System Is Less Efficient Than The Income Tax In Redistributing Income’, 23 The Journal of Legal Studies 667-681 (1994). E. Kempson, ‘Over-Indebtedness and its Causes Across European Countries’ in: H.-W. Micklitz and I. Domurath, Consumer Debt and Social Exclusion in Europe, (Farnham, Surrey, England; Burlington, VT, USA: Ashgate, 2015) 137-153. J. Köndgen, ‘Policy Responses to Credit Crises: Does the Law of Contract Provide an Answer?’ in: S. Grundmann and Y. Atamer (eds.), Financial Services, Financial Crisis, and General European Contract Law – Failure and Challenges of Contracting, (Alphen aan den Rijn: Kluwer, 2011), 35-59. J. S. Kraus, ‘Transparency and Determinacy in Common Law Adjudication: A Philosophical Defense of Explanatory Economic Analysis’, 93 Virginia Law Review 287-360 (2007). C. Kumpan and A. Hellgardt, ‘Haftung der Wertpapierdienstleistungsunternehmen nach Umsetzung der EU-Richtlinie über Märkte für Finanzinstrumente (MiFID)‘, 52 Der Betrieb (DB) 1714-1720 (2006). C. Lord, M. R. Lepper and E. Preston, ‘Considering the Opposite: A Corrective Strategy for Social Judgment’, 47 Journal of Personality & Social Psychology 1231-1243 (1984) J.-J. Laffont and D. Martimort, The Theory Of Incentives: The Principal-Agent Model, (Princeton, N.J.: Princeton University Press, 2002). D. Leczykiewicz and S. Weatherill (eds.), The Images of the Consumer in EU Law: Legislation, Free Movement and Competition Law (Oxford: Hart Publishing, 2016). P. Lunn, Regulatory Policy And Behavioural Economics, (Paris: Organization for Economic CoOperation And Development, 2014). European University Institute 43 Stefan Grundmann and Fabrizio Esposito W. A. Magat and W. K. Viscusi, Informational Approaches to Regulation (Cambridge, MA: MIT Press, 1992). W. A. Magat, ‘Information Regulation’, in: P. Newman (ed.), The New Palgrave Dictionary of Economics, Vol. II, (New York: Palgrave Macmillian, 2nd edn. 2002) 307-310. V. Mak, ‘Standards of Protection: In Search of the ‘Average Consumer’ of EU Law in the Proposal for a Consumer Rights Directive’, 1-2011 European Review of Private Law 25-42 (2011). V. Mak, ‘What is Responsible Lending? The EU Consumer Mortgage Credit Directive in the UK and the Netherlands’, 38 Journal Of Consumer Policy 411–430 (2015). H. Manne, ‘Mergers and the Market for Corporate Control’, 73 Journal of Political Economy 110-120 (1965). A. Marshall, Principles of Economics, (Palgrave Macmillian: London, 4th edn. 2013 (1920)). S. G. Medema, ‘Roll Over George Stigler: Berle And Means Meet Chicago Price Theory’, 6 Journal of Institutional Economics, 99-108 (2010). H.-W. Micklitz, ‘Introduction’, in: H.-W. Micklitz (ed.), The Many Concepts Of Social Justice In European Private Law (Cheltenham: Edward Elgar, 2011) 3-57. H.-W. Micklitz, ‘Conclusions: Consumer Over-Indebtedness and Consumer Insolvency – from Micro to Macro’, in: H.-W. Micklitz and I. Domurath, Consumer Debt and Social Exclusion in Europe, (Farnham, Surrey, England; Burlington, VT, USA: Ashgate, 2015) 229-235. A. Mian and A. Sufi, House of Debt: How They (and You) Caused the Great Recession, and How We Can Prevent It from Happening Again, (Chicago, London: University of Chicago Press, 2014) L. von Mises, Human Action. A Treatise On Economics, (San Francisco: Fox & Wilkes, 4th rev. edn. 1996 (1963)). W. Möschel, ‘Dogmatische Strukturen des bargeldlosen Zahlungsverkehrs’, 186 Archiv für die civilistische Praxis (AcP) 187-236 (1986). P. Nelson, ‘Information and Consumer Behaviour’, 78 Journal of Political Economy 311-329 (1970). R. S. Nickerson, ‘Confirmation Bias: A Ubiquitous Phenomenon in Many Guises’, 2 Review of General Psychology 175-220 (1998). OECD, Consumer Policy Toolkit, (Paris: Organization for Economic Co-Operation And Development, 2010). F. Parisi, The Language of Law and Economics: A Dictionary, (Cambridge: Cambridge University Press, 2013). J.E. Parkinson, Corporate Power and Responsibility: Issues in the Theory of Company Law (Oxford: Clarendon Press; New York: Oxford University Press, 1995). G. A. Papaconstantinou, ‘Investment Bankers in Conflict: The Regime of Inducements in MiFID II and the Member States’ Struggle for Fairness’, European Review of Contract Law (forthcoming, 2016). J. Persky, ‘Retrospectives: Consumer Sovereignty’, 7 The Journal of Economic Perspectives, 183-191 (1993). R. A. Posner, ‘Wealth Maximization Revisited’, 2 Notre Dame Journal of Law, Ethics & Public Policy 85-105 (1985). 44 Department of Law Working Papers Investor-Consumer or Overall Welfare R. A. Posner, ‘Are American CEOs Overpaid, And, If So, What If Anything Should Be Done About It?’, 58 Duke Law Journal 1013-1047 (2009). I. Ramsay, ‘Consumer Protection’, in: P. Newman, P. (ed.), The New Palgrave Dictionary of Economics and the Law, Vol. I, (New York: Palgrave Macmillian, 2nd edn. 2002), 410-415. I. Ramsay, ‘From Truth in Lending to Responsible Lending’, in: G. Howells, A. Janssen and R. Schulze, Reiner (eds.), Information Rights and Obligations, (Dartmouth: Ashgate, 2005) 47-65. D. W. Reekie, ‘Consumers’ Sovereignty Revisited’, 9 Managerial and Decision Economics, 17-25 (1988). N. Reich, ‘Economic Law, Consumer Interests and EU Integration’, in: H.-W. Micklitz, N. Reich and P. Rott (eds.), Understanding EU Consumer Law, (Portland; Oxford: Intersentia, 2009) 1-60. N. Reich, ‘Vulnerable Consumers in EU Law’, in: D. Leczykiewicz and S. Weatherill (eds.), The Images of the Consumer in EU Law: Legislation, Free Movement and Competition Law, (Oxford: Hart Publishing, 2016) 139-158. M. N. Rothbard, Man, Economy, And State: A Treatise On Economic Principles; With Power And Market, (Auburn, Ala.: Ludwig von Mises Institute, 2004 (1973)). P. Rott, ‘Consumer Credit’, in: H.-W. Micklitz, N. Reich and P. Rott (eds.), Understanding EU Consumer Law, (Portland; Oxford: Intersentia, 2009) 177-213. D. Rutherford, Routledge Dictionary of Economics, (New York: Routledge, 3rd edn. 2013). E. F. Scoles, E. C. Halbach and P. G. Roberts, Problems and Materials on Decedents’ Estates and Trusts, (New York: Aspen Publishers, 7th edn. 2006). J. C. Shepherd, The Law of Fiduciaries, (Toronto: Carswell, 1981). A.-L. Sibony and G. Helleringer, ‘EU Consumer Protection and Behavioural Sciences: Revolution Or Reform?’, in: A. Alemanno and A.-L. Sibony (eds.), Nudge And The Law, (Oxford: Hart Publishing, 2015) 209-233. A. Smith, An Inquiry into the Nature and Causes of The Wealth of Nations (Petersfield, Hampshire: Harriman House, 2007 (1776)). G. J. Stigler, ‘Competition’, in: D. Bulme and S. Durlauf (eds.), The New Palgrave Dictionary of Economics, (online edition, 2008). Study Group on a European Civil Code and the Research Group on EC Private Law (Acquis Group), Principles, Definitions And Model Rules Of European Private Law: Draft Common Frame Of Reference (DCFR), (München: Sellier, European Law Publishers, 2009). C. Teichmann, ‘Corporate Governance in Europa’, 2001 Zeitschrift für Unternehmens- und Gesellschaftsrecht 645-679 (2001). H. R. Varian, ‘Monitoring Agents with Other Agents’, 146 Journal of Institutional and Theoretical Economics 153–174 (1990). E. Vinter, A Treatise on the History and Law of Fiduciary Relationship and Resulting Trusts, (Cambridge: Heffer, 3rd edn. 1955). D. B. Wangrow, D. J. Schepker and V. L. Barker, ’Managerial Discretion: An Empirical Review and Focus on Future Research Directions’, 45 Journal of Management, 99-135 (2015). N. D. Weinstein, ‘Unrealistic Optimism About Future Life Events’, 39 Journal of Personality and Social Psychology 806-820 (1980); European University Institute 45 Stefan Grundmann and Fabrizio Esposito N. D. Weinstein and W. M. Klein, ‘Unrealistic Optimism: Present and Future’, 15 Journal of Social and Clinical Psychology 1-8 (1996). O. Williamson, ‘Managerial Discretion and Business Behaviour’, 53 The American Economic Review, 1032-1057 (1963). O. Williamson, ‘Corporate Finance and Corporate Governance’, 43 The Journal of Finance 567-591 (1988). S. Weatherhill, ‘Empowerment is not the Only Fruit’, in: D. Leczykiewicz and S. Weatherill (eds.), The Images of the Consumer in EU Law: Legislation, Free Movement and Competition Law, (Oxford: Hart Publishing, 2016) 203-221. The World Bank, Responsible Lending. Overview Of Regulatory Tools, (Washington DC, 2013). E. Wymeersch, ‘Unternehmensführung in Westeuropa – ein Beitrag zur Corporate GovernanceDiskussion’, 40 Aktiengesellschaft 299-316 (1995). E. Wymeersch, ‘Corporate Governance Codes and Their Implementation’, Festschrift for Horn, (Belin: De Gruyter, 2006) 619-631. W. Zöllner, Die Schranken mitgliedschaftlicher Stimmrechtsmacht bei den privatrechtlichen Personenverbänden, (Munich: Beck, 1963). F. Zuiderveen Borgesius, ‘Behavioural Sciences And The Regulation Of Privacy On The Internet‘, in: A. Alemanno and A.-L. Sibony (eds.), Nudge And The Law, (Oxford: Hart Publishing, 2015) 179-208. 46 Department of Law Working Papers