Ethics/Business

The Construction of a Trustworthy Investment Opportunity:

Insights from the Madoff Fraud*

HERV�E STOLOWY, HEC Paris

MARTIN MESSNER, University of Innsbruck

THOMAS JEANJEAN, ESSEC Business School

C. RICHARD BAKER, Adelphi University and Neoma Business School

1. Introduction

Economic exchanges rely to a great extent on information provided to and obtained by investors from different sources and channels. Economic theories of financial markets have long recognized the central role of information in animating markets (e.g., Arrow 1963; Ball and Brown 1968; Arrow 1984). Information is at the heart of any contractual rela- tionship. It allows investors to make “informed” investment decisions and to verify whether other contracting parties fulfill their obligations.

At the same time, engaging in a contractual relationship requires a certain level of trust. Trust is needed because the available information usually reduces only part of the uncertainty involved in a relationship. As Olsen (2008, 2189) puts it, “investors’ trust in the expertise and intentions of corporate managers, financial advisors and regulators is the ‘will of the wisp’ that creates and animates what we call the financial marketplace.” Recent literature in finance and economics has started to address the role of trust in finan- cial markets (Guiso, Sapienza, and Zingales 2008; Carlin, Dorobantu, and Viswanathan 2009; Bohnet, Herrmann, and Zeckhauser 2010; Sapienza and Zingales 2012). Some authors have considered the consequences that trust has for investors’ behavior. Guiso et al. (2008), for example, find that individuals who exhibit a high level of trust are more likely than others to invest in risky financial assets and tend to invest larger shares of their wealth in such assets. Holding constant the legal origin, Guiso et al. furthermore find that countries with a high level of trust exhibit higher stock market participation.

Other authors have examined the determinants of trust in financial markets and have pointed both to the characteristics of the financial system and characteristics of investors as factors that explain the level of trust. At the country level, the quality of institutions, such as regulatory bodies or the courts, has an important influence on the level of inves- tors’ trust (Zingales 2009). At the individual level, Alesina and La Ferrara (2002, 231) find

* Accepted by Yves Gendron. The authors express their gratitude to all their interview partners and, espe-

cially, to Madoff’s investors who accepted, despite their pain and often financially distressed situation, to be

interviewed for this research. The authors gratefully acknowledge comments by Diane-Laure Arjali�es, Thi-

erry Foucault, Yves Gendron, Chris Humphrey, Lambert Jerman, Nancy Leo, Sabina du Rietz, two anony-

mous reviewers, participants at the EAA Annual Meeting (Rome, April 2011) and AFC Annual Meeting

(Montpellier, May 2011), and workshop participants at York University (October 2010), Paris Dauphine

University (February 2011), the University of Manchester (February 2011) and the University of Bristol

(March 2011). Responsibility for the ideas expressed, or for any errors, remains entirely with the authors.

Thomas Jeanjean and Herv�e Stolowy acknowledge the financial support of the European Commission (INTACCT project, contract No. MRTN-CT-2006-035850). Herv�e Stolowy is a member of the GREGHEC,

CNRS Unit, UMR 2959.

Contemporary Accounting Research Vol. 31 No. 2 (Summer 2014) pp. 354–397 © CAAA

doi:10.1111/1911-3846.12039

that trust is a function of individual characteristics (such as education, gender, or income) as well as of the characteristics of the community from which the investor comes (such as the level of income inequality).

This literature has advanced our understanding of the role of trust in important ways. Yet, in focusing on individual- and country-level factors, prior studies have not paid much attention to the particular “objects of trust.” What are the mechanisms through which a particular investment opportunity is made to appear trustworthy? What is needed to convince investors that a given financial product can be trusted? Existing lit- erature does not provide a detailed analysis of the process through which trustworthiness is established in financial markets. In our paper, we therefore seek to address this gap in the literature. We inquire into the production of trust in the context of financial markets by examining the process through which trustworthiness is constructed in the eyes of investors.

The role that trust plays in social and economic life may easily be overlooked, due to its implicit nature (Giddens 1990; Sztompka 1999). Somewhat paradoxically, the influence of trust on our decisions and actions is perhaps most visible in those cases where trust turns out to be unwarranted. In the particular context of financial markets, this may be the case when expectations regarding the functioning and regulation of such markets or regarding the potential return on a certain investment opportunity are not confirmed by reality. Investors then face negative consequences from their trusting behavior and, with the benefit of hindsight, they realize that “they should not have trusted” to the extent they did.

Our paper draws upon empirical evidence from one such incident in which the fragile nature of trust was revealed in a rather dramatic way. We analyze the role of trust in the spectacular investment fraud of Bernard Madoff, which was exposed at the end of 2008. Madoff had for many decades run a wealth management business, Bernard L. Madoff Investment Securities LLC (BMIS), without investing his clients’ money in securities. He created a so-called Ponzi scheme

1 , where money from new investors is used to pay interest

and dividends to existing ones. The “success” of such a scheme depends on the trust that investors have in the particular investment opportunity and/or its propagators. In order to understand the mechanisms behind the development of such trust, we conducted inter- views with individual U.S. investors who lost money in the Madoff fraud. We complement the interview material through the analysis of letters written by investors (victim-impact statements) and through other publicly available information about Madoff and his activi- ties. What emerges from our analysis of this “extreme case” (Flyvbjerg 2001) is an interest- ing set of insights into how the investment opportunity proposed by Bernard Madoff came to be seen as a trustworthy one. To organize these insights, we draw upon estab- lished concepts and arguments regarding the production of trust. In particular, we rely on the distinction between process-based, characteristic-based, and institution-based trust (Zucker 1986; Neu 1991b, 1991a),

2 which we enrich by incorporating other important

ideas from the sociology and accounting literatures (Sztompka 1999; Tomkins 2001; Barrett and Gendron 2006).

We see the main contribution of our paper in the way in which it illuminates the mechanisms behind the production of trust in the context of financial markets. To our knowledge, our paper is the first to analyze in rich empirical detail how different forms of trust help to construct an investment opportunity as trustworthy. Our analysis

1. In a Ponzi scheme, existing investors are paid purported returns out of the funds that new investors contrib-

ute. “Ponzi schemes tend to collapse when it becomes difficult to recruit new investors or when a large num-

ber of investors ask to cash out” (U.S. Securities and Exchange Commission 2010).

2. Zucker (1986) and Neu (1991b, 1991a) speak of “institutional-based” trust. We prefer to use the more com-

mon wording “institution-based.”

Construction of a Trustworthy Investment 355

CAR Vol. 31 No. 2 (Summer 2014)

demonstrates not only the role of the intensity of the relationship between Madoff and his investors, but also that of the consistency of the signals obtained by the investors through more impersonal channels: consistency in time (through the impressive track record that Madoff claimed) and in space (through the association with many different prestigious institutions). This consistency, we argue, created an illusion of trustworthiness that inves- tors were apparently unable to see through. Reflecting more generally on the nature of the “reality” of financial markets, we suggest that some element of illusion is inherent to the functioning of such markets, which might explain part of the difficulty of detecting a financial scam. Our paper therefore contributes also to the discussion about “hyperreal” financial markets (McGoun 1997; Macintosh, Shearer, Thornton, and Welker 2000) by highlighting the link between mechanisms of trust creation, on the one hand, and concerns with self-referential representations, on the other hand. In particular, we identify similari- ties between the trust dynamics in the Madoff case with those observed in the literature on speculative bubbles. More generally, our examination of the construction of trust in financial markets resonates with recent calls for, and examples of, a sociology of finance and financial markets (e.g., Callon 2009; MacKenzie 2009; Vollmer, Mennicken, and Preda 2009).

The remainder of the paper is organized as follows. In the next section, we elaborate on the theoretical concepts that guide our analysis. The third section describes our research approach and methods. After providing some background information on the empirical context of our study in the fourth section, we analyze the various mechanisms that contributed to the production of trustworthiness in the Madoff case. The discussion that follows draws together our findings and the conclusions that may be drawn from them. The final section concludes with some avenues for future research.

2. Theoretical background

Sources of trust

Economic decisions are based on various types of information, but rarely is such informa- tion “complete.” Usually, information reduces only part of the uncertainty in the environ- ment. The remaining gap is frequently filled through trust. As Tomkins (2001, 165) says, “trust implies adopting a belief without full information.” For example, when making financial investments, investors may choose not to obtain additional information about the firms in which their investment advisor places their money, because they trust that the advisor will make the right decisions. While trust is thus “an alternative uncertainty absorption mechanism to increased information” (Tomkins 2001, 165–66), trust also relies on information. There must be some reason for trust to develop, and this reason is linked to available information about the particular target of trust.

Trust may be oriented toward different targets: we can have trust in individual per- sons, in groups of people, in organizations, in technologies, in state institutions, etc. The basic logic of trust at work here is in principle always the same (Sztompka 1999, 46): trust implicitly means trust in people and their actions. For example, if we “trust an organiza- tion,” we implicitly trust the people who manage or are employed in that organization. If we “trust a technology,” we implicitly

3 trust the people who have designed and con-

structed this technology or perhaps also those who have tested and certified it or even those who testify having successfully used it, for example.

What are the types of information that turn potential targets of trust into trustwor- thy ones? What are, in other words, the sources of our trust? Zucker (1986) suggests that

3. We say “implicitly” here to account for the existence of trust in systems (Giddens 1990). When people invest

trust in systems, they do not necessarily think about the people who have designed the systems or who

operate them, although they implicitly trust such people when they trust the system.

356 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

trust in the actions of another person may be based on three types of information: infor- mation about past exchanges with that person, such as her reputation; information about the characteristics of the person, such as her family background, age, or ethnicity; or information about certain institutions and their functioning, such as certificates, profes- sional bodies, or educational degrees. Accordingly, Zucker distinguishes between process- based trust, characteristic-based trust, and institution-based trust. Although these sources of trust may often overlap in practice, their distinction seems helpful for analytical purposes.

In the case of process-based trust, the key source of information about the trustwor- thiness of the person is information about past exchanges with that person. There exists, in other words, a history of past actions from which we extrapolate into the future. Some- times, we can resort to our personal observations and memories in order to construct such a history, such as when we have had prior dealings with a business partner or when we can look back at a long-term personal relationship. In other cases, we need to use “secondhand testimonies” about the conduct of others, which we may obtain, for example, from the media or from relatives and friends (Sztompka 1999, 72). Whatever the source of our knowledge, trustworthiness is constructed on the basis of some information about the history of a person’s achievements and behavior.

Characteristic-based trust is different insofar as it relates to more general characteris- tics of a person, such as nationality, gender, or occupation (Zucker 1986). Such character- istics can create trustworthiness both within and across groups, as pointed out by Neu (1991a). If people share the same characteristics, this creates a common background within the group, which reduces the perceived need to negotiate the terms of exchange or to inquire into the other person’s credibility (Zucker 1986, 61). Across groups, such common background does not exist, but knowledge about the characteristics of the other person also creates certain expectations about their behavior, as is the case when we “stereotype” other people’s behavior on the basis of gender or nationality (Neu 1991a).

The third type of trust that Zucker (1986) distinguishes is institution-based trust. Here, people rely on the work of institutions, such as laws and regulations, certificates, professions, or educational systems, to ensure the proper functioning of social or economic exchanges. The existence of such institutions produces comfort and reduces the perceived need to monitor personally or control a given exchange process.

Neu (1991a, 1991b) suggests distinguishing between three classes of institution-based trust. The first class relates to individual and firm-specific actions that involve the acquisi- tion or adoption of certain institutional forms such as educational degrees, certificates, or “best practice” tools. In acquiring or adopting these forms, a person or organization can signal expertise or legitimacy and, thus, trustworthiness (see also Meyer and Rowan 1977). A second class comprises intermediaries that provide some sort of warranty or guarantee for the functioning of the relationship even when the exchanging parties have had no prior exposure to each other. Auditors are a case in point. Auditing firms are supposed to verify the accounting information provided by firms and, in so doing, enhance trust in the partic- ular relationship (Neu 1991a). Regulation constitutes a third class of institution-based trust. Regulatory bodies, such as the Securities and Exchange Commission (SEC) or the European Commission, seek to make the behavior of actors more predictable through a set of prescriptions that ought to be followed and by instituting monitoring and control mechanisms through which rule-following purports to be enforced (ibid.). The trustworthi- ness that these bodies generate is, to an important extent, based on accountability require- ments that they create.

The different sources of trust outlined in this section imply that people form certain expectations about the future. The exact strength of these expectations may vary, however, as discussed next.

Construction of a Trustworthy Investment 357

CAR Vol. 31 No. 2 (Summer 2014)

Strength of expectations

When a person invests trust in another person or organization, she expects this other party to behave in a certain way. Expectations can vary in terms of their strength (Neu 1991a, 1991b). At one end of the continuum, perceived trustworthiness is very low, such that one hardly expects the other party to meet the agreed-on obligations and perhaps already expects some level of wrongdoing. Nevertheless, one has sufficient trust so as to engage in the relationship, or perhaps one needs to engage in it because of a lack of alternatives. At the other extreme, trust is so high that one can hardly imagine any wrongdoing and there- fore expects instead that the other person will exercise particular care and benevolence.

Process-based, characteristic-based, and institution-based trust can, in principle, all create strong expectations. Yet, there are some factors that make it more likely that strong expectations will in fact develop. In the elaborations that follow, we combine insights from different authors (Neu 1991a, 1991b; Sztompka 1999; Tomkins 2001; Barrett and Gendron 2006) who have all discussed this point at some length.

Generally speaking, the more reasons one has to trust someone else, the stronger one’s expectations regarding that person or institution will be. “More” can therefore mean dif- ferent things. We suggest that these reasons can be categorized into a question of time; a question of space; and a question of intensity.

Time is strongly associated with consistency of conduct: “the better and longer we are acquainted with somebody, and the more consistent the record of trustworthy conduct, the greater our readiness to trust” (Sztompka 1999, 72). This is particularly apparent for process-based trust that is obtained firsthand (i.e., through a personal history of exchange). However, even if process-based trust is obtained secondhand (i.e., by relying on the testimony of others), the time dimension plays a role. For instance, repeated reports about the good performance of a firm are likely to increase our trust in the firm as an investment object. Time is also relevant for the strength of institution-based trust. Expectations will be higher when there are more positive firsthand experiences with a par- ticular institution. Trust will also increase if we learn secondhand about the lengthy exis- tence and positive track record of an institution. A well-established MBA degree, for example, creates higher expectations regarding the abilities of its graduates than a recently created one.

The idea that trust can increase over time has an important implication for our under- standing of the relationship between trust and information, as illustrated by Tomkins (2001) in the case of business relationships. Depending on how well established a relation- ship is, the amount of trust that exists and the amount of information that is required will differ. At the beginning of a relationship, the level of trust will be relatively low, especially if there is also little secondhand information available. As the relationship matures, one accumulates more information about the other party’s behavior, and the level of trust is therefore likely to increase. In the later stages of the relationship, less information will be needed to sustain the trust that has been built up because there is already a sufficiently large stock of positive experiences (Tomkins 2001, 170). As a result there is an inverse U-shaped relationship between the perceived need for additional information (or control mechanisms) and the level of trustworthiness of the other party.

Not only does consistency in conduct over time increase trustworthiness; consistency in space also increases trust. If a potential business partner does not have a long-term track record, we may still have stronger expectations regarding his behavior if we know that he is currently also dealing with other reputable firms (Sztompka 1999, 73). The more we perceive someone to be part of a legitimate network, the more this person will benefit from the trustworthiness of the other network members. Similarly, we are more likely to associate particular characteristics, such as religion or gender, with benevolent behavior if

358 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

we know many (rather than few) people with such characteristics who also demonstrate trustworthy conduct. Furthermore, in the case of institution-based trust, consistency in space can make a difference regarding the level of expectations. A particular institution will gain in trustworthiness if it is associated with many other trustworthy institutions or persons that can lend it legitimacy. For example, we are more likely to trust the quality of a particular university degree if that degree is offered by more (rather than fewer) presti- gious universities. In each of these cases, there is a sort of “bandwagon effect” at work (Sztompka 1999, 73) in the sense that “the fate of an object of trust … is tightly related to trust vested in other objects” (Barrett and Gendron 2006, 634).

Finally, the strength of one’s expectations vis-�a-vis others is also a function of the intensity of one’s experiences. This point is particularly stressed by Neu (1991a, 1991b) in his discussion of trust-generating mechanisms in the stock market. Neu suggests that the more personal or intense a relationship is, the stronger one’s expectations toward the other person will be. This is especially the case when trust is established firsthand, such as in a close business partnership or in family bonds. In a very close relationship, one expects the other party not only to fulfill their contractual or legitimate obligations but also to “honor” the relationship; that is, to enact a certain sense of personal obligation that goes beyond purely self-interested behavior (Neu 1991a, 187). In the case of a business relation- ship, strong expectations introduce a “social override” into what would otherwise be a purely economic relationship. Neu (1991a) suggests that process-based experiences that are generated firsthand tend to be more intense in this respect than secondhand ones, which is why trust is likely to be stronger in the first case. They are also likely to be more intense than characteristic-based or institution-based mechanisms, which are more impersonal by definition. Institutions may not evoke feelings of social closeness or intimacy, even if they are well-established (Neu 1991a). In fact, intermediaries and regulators usually try to delineate what can and cannot be expected of them. They establish particular forms of control and accountability that “determine the strength of the expectations along with the situations in which the expectations should and should not apply” (Neu 1991a, 189).

4

Hence, to sum up the idea of intensity as a determinant of the strength of expectations, we can say that intimate relationships produce a stronger expectation that the other party will do more than merely meet their legitimate obligations. Conversely, in less intimate relationships, trust is (at best) restricted to the comparatively weaker expectation that the other party will fulfill its legitimate obligations.

So far, we have established that trust emerges in the form of process-based, character- istic-based, and institution-based trust and that the level of expectations in each case will vary with the quantity and quality of the available cues regarding trustworthiness, as defined through the dimensions of time, space, and intensity. We will use this conceptual framework to examine the production of trustworthiness in the Madoff fraud. Before doing so, we now discuss our research methodology.

3. Research approach and methods

Our inquiry into the dynamics of trust in financial markets is based on a case study of one particular incident that created a great deal of attention in the media and among the general public: the investment fraud of Bernard Madoff that was eventually revealed in 2008. While this case arguably constitutes an “exceptional” event in many respects, we would like to avoid viewing this case only or even primarily in terms of its exceptional nature. Rather, we believe that a notorious case like the Madoff case can provide

4. This does not mean, of course, that people’s trust is always restricted in this way. It may well be the case

that people expect institutions to do more than they actually claim to do. This is not a question of the

intensity of the relationship, but rather of the social construction of expectations.

Construction of a Trustworthy Investment 359

CAR Vol. 31 No. 2 (Summer 2014)

important insights into the functioning of financial markets more generally. Indeed, it is often with the help of “extreme cases” (Flyvbjerg 2001) that we can better understand some basic mechanisms that are of general relevance but are difficult to discern in “aver- age” cases, where they appear in less visible forms. Cooper and Morgan (2008) emphasize the potential of extreme cases to further our understanding of accounting phenomena, and they provide several examples of studies that have pursued such a research strategy.

In our investigation of the Madoff fraud, we adopt a qualitative research approach. This involves the use of qualitative data (text) as well as the analysis of such data in a qualitative (interpretive) manner (Silverman 2004). Our data come from two sources: inter- views with individuals who invested with Bernard Madoff and letters written by investors in support of Madoff’s judicial sentencing. A qualitative approach seems appropriate for our research interest for at least two reasons. First, in order to understand the dynamics of investment decisions, it is important to consider actors’ own beliefs and perceptions about the situation and its context, rather than an outsider’s description. For example, in order to understand why an investor does or does not invest money in a fund, it is not rel- evant whether the fund is or is not well managed according to some objective standards. What counts is whether the investor believes the fund is well-managed or not. Qualitative methods are helpful in uncovering such beliefs; that is, “for understanding the world from the perspective of those studied” (Pratt 2009, 856). Previous accounting researchers have used qualitative methods in similar empirical settings (e.g., Gendron and Spira 2009). Sec- ond, the particular case of investment fraud constitutes a “delicate” topic to investigate. Given the enormous financial losses and personal setbacks that many of the investors suf- fered, it was important to relate to them with empathy, which is arguably easier in a per- sonal conversation even if “only” by telephone or voice over the Internet than through impersonal means such as questionnaires.

In order to identify persons who invested with Madoff, we consulted several sources. In February 2009, the U.S. federal bankruptcy court released a full list of Madoff’s clients (available through the Internet).

5 This list is 162 pages long and lists approximately 14,000

investors. Some of the investors are mentioned multiple times, presumably because they had more than one account with Madoff. According to Sander (2009, 229), the actual number of individuals who invested with Madoff was 11,374, across 44 U.S. states and 44 countries, with the majority concentrated in the New York and Florida areas.

More detailed information regarding some of the investors was obtained from letters sent by the investors in support of Madoff’s sentencing. In June 2009, federal prosecutors in New York City filed statements from 113 alleged victims of Bernard Madoff’s fraud with the U.S. district court judge.

6 The filing was made prior to Madoff’s sentencing on

June 29, 2009. These letters, or “victim-impact statements,” ranged from one paragraph to several pages, and they contain personal stories of the investors and their losses. The 113 statements can be subdivided into two categories: 65 “direct” investors (including eight “direct” investors who request to speak at the sentencing) and 48 “indirect” investors

7

who ask for the right to be considered in the same way as direct investors.

5. Sources: http://www.businessinsider.com/2009/2/bernie-madoffs-clients-the-official-list (last retrieved: Febru-

ary 1, 2013); http://s.wsj.net/public/resources/documents/st_madoff_victims_20081215.html (last retrieved:

February 1, 2013; http://richard-wilson.blogspot.com/2009/01/bernard-madoff-fraud-victims-list-2008.html

(last retrieved: February 1, 2013). See also Sander (2009, 249, Appendix D) for a list which shows the inves-

tor type (e.g., feeder fund, hedge fund, bank, charity, family office, individual).

6. These statements are available from many sources (e.g., http://www.cnbc.com/id/31375911/

Read_Them_Here_Madoff_Case_Victim_Statements) (last retrieved: February 1, 2013). A few of these state-

ments have been written by former employees of Madoff’s legitimate activity.

7. Indirect investors are thus called because they invested indirectly: “through a bank, a mutual fund, or an

arm’s length ‘feeder’ fund set up specifically to direct money to Madoff without investors’ knowledge”

(Arvedlund 2010, ii).

360 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

We decided to examine these 113 statements for two reasons. First, we thought that the investors who narrated personal stories in written form would be the ones most willing to be interviewed. Second, reading the letters provided us with some background informa- tion regarding the investors, such as whether they had invested directly or indirectly in the Madoff funds. This information proved helpful in preparing our interviews.

Among the 113 statements, 45 included an email address while 12 included only a postal address or telephone number. Accordingly, we sent out 45 emails and 12 letters ask- ing for an interview: 55 contacts were made with direct investors (out of 65), and two emails were sent to indirect investors: one who represented 46 other indirect investors in a “Ponzi victims coalition” and one who contacted the judge directly. Eleven investors (10 following emails and one after having received a letter) accepted an invitation to have a conversation with us. Given that the investors who agreed to be interviewed were located throughout the United States, we decided to conduct each interview by telephone or voice over Internet by the same author for consistency purposes. We followed the interview guide shown in Appendix 1. We asked the same questions in all interviews, but we also allowed for additional questions that would make it possible to delve deeper into our respondents’ answers. The interviews were recorded with the approval of the interviewees and were subsequently transcribed.

Given the stressful situation for most of our interviewees, we considered it important to try to establish a trustworthy relationship with them. Lincoln and Guba (1985, 303) suggest that trustworthiness needs time to develop and that researchers should therefore strive for a “prolonged engagement” with their informants. Such prolonged engagement is obviously difficult to achieve when talking to someone for the first time. In order to miti- gate this problem, we devoted the first few minutes of each conversation to introducing ourselves and laying out our research interest. We carefully explained the objective of our research, indicated that the research team included only academics and emphasized that we intended to publish an academic article rather than a press article. The immediate feed- back we received from our interviewees was positive and made us rather confident that the interviewees would provide us with truthful accounts of their experiences. Indeed, dur- ing the interviews, we had no reason to believe that the interviewees manipulated their answers so as to manage impressions or gain legitimacy (Alvesson 2003).

Among the 11 investors interviewed, nine were direct investors in Madoff’s funds and two (Investors 6 and 7) invested indirectly through a feeder fund. Two main questions guided our interviews: What led people to invest with Madoff, and how did this impact their behavior prior to the investment decision? What kind of information did investors receive during the investment period, and how did this influence their behavior along the way?

In addition to the interviews, we analyzed in detail the statements written by the 65 direct investors and found, in 26 instances, some textual material related to our research topic and corresponding to our interview guide.

8 Among these 26 statements, eight were

written by investors that we interviewed. Consequently, to avoid double counting, we removed these statements from our dataset. Our resulting sample includes 29 investors: 11 interviews and 18 statements from noninterviewed investors.

9

Data analysis was carried out by reading through the interview transcripts and the impact statements and by moving back and forth between data and theory until we were

8. The 39 other statements have been discarded because they did not contain any information regarding the

investment process as such and thus not address the construction of trust. These statements typically pro-

vided only an account of the losses incurred by the investor and of the negative impact that these losses

had on their personal lives.

9. Interviewed investors are referred to as “Investor 1 to 11” and the other investors are referred to as “Inves-

tor 12 to 29.”

Construction of a Trustworthy Investment 361

CAR Vol. 31 No. 2 (Summer 2014)

able to identify patterns that we thought provided interesting theoretical insights (Ahrens and Chapman 2006). In interpretive research, the validity or trustworthiness of a study’s findings can be operationalized through the concepts of authenticity and plausibil- ity (Lukka and Modell 2010). We sought to achieve authenticity by taking the views of our respondents seriously and by letting our respondents speak, with the help of direct quotations, throughout the resulting narrative. We thereby sought to balance the “telling” and “showing” elements of our empirical analysis (Golden-Biddle and Locke 2007). As far as plausibility is concerned, we discussed the potential of alternative theoretical concepts to explain our empirical material, which was facilitated by the diversity (in terms of disciplinary background) within the team of researchers as well as by the feedback received from the editor, the two reviewers, and other academic colleagues.

10

One risk of using retrospective interview reports is that interviewees may rationalize their behavior and experiences after the fact. Such post hoc rationalization can have cognitive reasons, such as when interviewees cannot recall each and every detail of what they have done (Ericsson and Simon 1980), or it may be due to motivational reasons, such as when they report in a self-serving manner (Heider 1958). Although the existence of post hoc rationalization cannot be completely eliminated in retrospective interview studies, we have tried to limit its magnitude and influence in different ways. First, we sought to reduce the influence of individual recall biases by interviewing several different investors and by working with the patterns that we identified across their responses (i.e., triangulation between interviews). Second, we followed Huber and Power’s (1985) sugges- tion to focus on very factual questions and to avoid eliciting judgments with our ques- tions. Arguably, this is especially important when interviewees have a high emotional involvement in what happened, as they may then be prone to frame their narratives in light of their emotions (Huber and Power 1985, 175). Finally, we avoided wording our questions in a way such that interviewees would be tempted to answer in the positive. For example, we never directly asked our interviewees to comment on the relevance of trust in their decision making but merely requested them to recount the motivation for their investment decision.

We complemented the interviews and statements with several other sources of data that allowed us to obtain a better understanding of the fraud and of the relationship between Madoff’s company and the investors. First, we conducted two further interviews: one with the chief financial officer (CFO) of an organization that is close to the Jewish community and was mentioned in the media because it chose not to invest with Madoff (Strober and Strober 2009, 42)

11 and one with the chief executive officer (CEO) of a com-

pany specializing in defending the interests of minority shareholders, who represents sev- eral of Madoff’s investors.

In addition to our interviews, we also consulted publicly available primary mate- rials. In order to see how Bernard Madoff was constructed as a trustworthy invest- ment manager in the media, we retrieved press articles through the Factiva database.

12 We searched the period from January 1, 1960 (when Madoff started his

10. As noted by one of our reviewers, the production of a scientific work involves the production of trustwor-

thiness with the help of mechanisms similar to the ones described in our paper for the case of financial

investments. Here and there, it is up to the readers (investors) to judge whether the output is sufficiently

trustworthy. The production of trustworthiness in the scientific realm has been particularly discussed by

Latour (1987).

11. In the remainder of the paper, quotations from this interview will be made by reference to

“noninvestor 1.”

12. Factiva is a nonacademic database of international news containing 20,000 worldwide full-text publications

including the Financial Times, and the Wall Street Journal, as well as the continuous information from

Reuters, Dow Jones, and the Associated Press (see http://www.dowjones.com/factiva) (last retrieved:

February 1, 2013).

362 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

activity) to November 30, 2008 (a few days before the announcement of the fraud, see Table 1).

14 As we were interested in the potential impact that such press articles

could have on the views of investors, including the nonsophisticated ones, we decided to exclude specialized professional journals and magazines and to use the general press instead. More specifically, we searched the 25 newspapers in the United States with the highest circulation (i.e., Wall Street Journal, USA Today, New York Times, etc.).

15 We

conducted a search on the keywords “Bernard Madoff” and “Bernie Madoff.” 16

This search resulted in 47 articles, each of which we carefully read to identify instances of the construction of trustworthiness.

TABLE 1

Chronology

1938, April 29 Bernard Madoff born in New York City. 1960 Madoff graduates from Hofstra College, Hempstead, New York, with a degree

in political science. He qualifies as a general securities representative

(salesperson) and general securities principal (allowing him to establish a securities sales and trading firm). The securities firm of Bernard L. Madoff Investment Securities is founded and registered with the SEC as a broker-dealer.

1962 Accountants Frank Avellino and Michael Bienes begin to raise investments for Madoff, promising returns of 12 to 20 percent.

1967 Bernard Madoff’s brother, Peter, graduates from Fordham Law School and

joins his brother’s firm as a partner. 1971 The NASDAQ over-the-counter securities trading market is created. 1983 Madoff Securities International opens in London. 1984 Madoff becomes a member of the Board of Governors of National Association

of Securities Dealers (securities self regulatory organization). 1985 Cohmad Securities is founded (the legitimate part of Madoff’s securities trading

business).

1990 Madoff becomes Chairman of NASDAQ. 1992 Avellino and Bienes are accused of illegally selling securities to thousands of

investors over three decades. Madoff refunds the money.

1993 Avellino and Bienes are forced to shut down and pay a fine. 2000 Harry Markopolos submits his first allegations concerning the Madoff fraud to

the SEC in Boston. 2001 Stories about Madoff appear in the financial press questioning Madoff’s success

(see Arvedlund 2001; Ocrant 2001). 2005 “The World’s Largest Hedge Fund Is a Fraud” (Harry Markopolos).

13

2006 SEC opens an investigation of Madoff.

2006 Madoff registers as an investment adviser. 2007 SEC closes investigation of Madoff, finding no fraud. 2008, Dec. 10 Madoff confesses to his brother and his sons.

2008, Dec. 11 He is arrested by the FBI. 2009, March 12 Madoff pleads guilty to 11 criminal charges. 2009, June 29 He is sentenced to 150 years in federal prison. 2010, December 11 Suicide of Mark Madoff, Bernard Madoff’s eldest son.

13. Available at: http://static.reuters.com/resources/media/editorial/20090127/Markopolos_Memo_SEC.pdf (last

retrieved: February 1, 2013).

14. Two detailed chronologies have been published (see Kirtzman 2009, 273–79; Sander 2009, 255–59). 15. http://en.wikipedia.org/wiki/List_of_newspapers_in_the_United_States (last retrieved: February 1, 2013).

16. It was impractical to make a research on the name “Madoff” because too many results turned out to be

unrelated to Bernard Madoff.

Construction of a Trustworthy Investment 363

CAR Vol. 31 No. 2 (Summer 2014)

Finally, we consulted other materials, such as videos of victims published on the Inter- net

17 as well as books (Arvedlund 2009; Kirtzman 2009; LeBor 2009; Oppenheimer 2009;

Ross 2009; Sander 2009; Strober and Strober 2009; Weinstein 2009; Arvedlund 2010; Sar- na 2010; Henriques 2011) and press articles (see, e.g., Seal 2009) that addressed the Mad- off fraud. It should be pointed out that it is difficult to verify the information provided in secondary sources, especially in the case of popular books; one has to take into consider- ation a certain level of “journalistic creativity” through which a compelling narrative is constructed. As a consequence, we cannot rule out the possibility that some of the infor- mation provided in the secondary sources is strongly framed or misrepresented. However, given that we use these sources mainly for creating a background for understanding of the case rather than to inform our theorizing directly, we believe that we have mitigated the impact of such problems. In those cases where we quote interview sequences provided in secondary sources, these quotations serve to support evidence that we collected firsthand in our own interviews.

In the following sections, we present the analysis of the empirical material, organized around different mechanisms that contribute to the production of trustworthiness. Before doing so, we provide some summary background about the Madoff fraud.

4. The investment fraud of Bernard Madoff

Bernard Madoff, who was born and raised in New York City, founded Bernard L. Madoff Investment Securities LLC (BMIS) in 1960. He became an important player in the stock markets and then a pioneer and leader in building the NASDAQ stock exchange (Sander 2009, 37–39). BMIS was one of the top market makers on Wall Street (Lieberman, Gogoi, Howard, McCoy, and Krantz 2008).

Madoff remained chairman of the board of directors of BMIS until his arrest on Decem- ber 11, 2008 (Voreacos and Glovin 2008). As his business grew, Madoff was eventually able to purchase numerous properties including houses, apartments, and boats in locations around the world. However, he was said to have lived in a relatively unostentatious manner (Arvedlund 2009, 13; Strober and Strober 2009, 23). According to a government filing in March 2009, Madoff and his wife had a net worth of about $126 million, plus an estimated $700 million comprising the value of his interest in BMIS (McCool and Honan 2009).

On December 10, 2008, Madoff’s sons informed federal authorities that their father confessed to them that the investment management part of his firm was fraudulent and quoted him as saying it was “one big lie” (Appelbaum, Hilzenrath, and Paley 2008; Henri- ques 2011). The following day, FBI agents arrested Madoff and charged him with securi- ties fraud. In March 2009, he pleaded guilty in the United States federal district court in New York City to 11 federal crimes, and he admitted that his wealth management busi- ness was a Ponzi scheme.

Madoff claimed to use a split-strike strategy. A split-strike strategy consists of acquir- ing a basket of stocks highly correlated to the S&P 100 index as well as call and put options to create a ceiling and a floor for the gains and losses from the acquired stocks. In theory, this strategy allows managing risks while taking advantage of the high return from the acquired stocks (Bernard and Boyle 2009). However, according to Ross (2009, 185), Bernard Madoff perpetuated his fraudulent scheme by never making any trades for his investors. On a daily basis, Frank DiPascali, Madoff’s chief financial officer, kept track of the closing prices of the S&P 100. Then, on a regular basis, DiPascali and Madoff would pick the stocks that had done well and create false trading records for their “basket of stocks” (Ross 2009, 74). One of his employees was in charge of producing statements

17. See, e.g., http://www.vanityfair.com/politics/features/2009/04/madoff200904?printable=true (last retrieved:

February 1, 2013) or http://www.youtube.com/watch?v=U3iseHqUzak (last retrieved: February 1, 2013).

364 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

reflecting thousands of trades that were supposedly being executed for Madoff’s invest- ment clients (Ross 2009, 73).

Madoff confessed that he had begun his fraudulent financial scheme in the early 1990s. However, federal investigators believe that the fraud began as early as the 1980s (Ross 2009, 205; Safer 2009). While the amount missing from client accounts is alleged to be as much as $65 billion, the court-appointed trustee has estimated the losses to investors were actually about $18 billion (Henriques 2009a). On June 29, 2009, Madoff was sentenced to 150 years in prison (Frank, Efrati, Lucchetti, and Bray 2009).

The collapse of Madoff’s investment company and the subsequent freezing of his assets and those of his firm affected businesses, charities, and foundations around the world. In Table 1, we present a summary chronology of the main events of the Madoff case.

5. The construction of trustworthiness in the Madoff case

Our review of the literature on trust has revealed that trust needs some reason to develop, and this reason will often relate to the perceived trustworthiness of persons or institutions. This raises the question how the investment opportunity that Madoff offered came to be seen as trustworthy. To address this question, we look first at the motivations that inves- tors give for why they decided to invest their money with Madoff. These accounts reveal important sources of trust which, in turn, are informative about the mechanisms through which trustworthiness is constructed.

When asked about how they got into contact with Madoff’s investment fund, several of our interviewees referred to other persons, such as relatives, friends, or business part- ners, who apparently recommended investing with Madoff. The following two excerpts provide an illustration:

I had a partner on the account, and my partner’s brothers worked for a firm on Long

Island, in New York, who had their retirement plans with Madoff. And we heard about

it through that connection. [My partner’s] brother worked for that company and was

telling us how pleased he was with his profit-sharing plan with the company. (Investor

2, interview)

My mother was a direct investor.… I originally went in on that account to help her,

because I am a professional.… Friends recommended that she go to this. (Investor 9,

interview)

It is often normal for people to rely on information received from others when making a decision. Most first-time experiences or exchanges with another person or party are based on information that is obtained secondhand, given that there is no firsthand experience available. Depending on the trustworthiness of the person who provides the information, the investment opportunity will appear to be more or less trustworthy.

At the same time, even the most trustworthy friends will hardly be trusted “blindly.” When important amounts are at stake, most people will ask for at least some informa- tion about the particular investment opportunity. Such information provides an addi- tional source of trust, over and above the trust in relatives or friends. Indeed, the investors we interviewed made reference to different types of information that they had received or collected about Madoff. In this respect, the distinction between process- based, characteristic-based, and institution-based trust (Zucker 1986; Neu 1991a, 1991b) is helpful for understanding the different ways in which trustworthiness can be constructed.

Construction of a Trustworthy Investment 365

CAR Vol. 31 No. 2 (Summer 2014)

Performance and reputation: Process-based trust

A first form of trust that emerged from our interviews is what Zucker refers to as process- based trust; that is trust “tied to past or expected exchange such as in reputation or gift- exchange” (Zucker 1986, 53–54). What counts here is the fact that there is some past track record that fuels one’s expectations about the future. The tendency to make investment decisions on the basis of past information is well documented in empirical research (Wilcox 2003)—despite the questionable rationality of such behavior from an economic point of view.

As Neu (1991b, 247) explains, “individuals rely on transaction-specific information to infer that the necessary trust exists for an exchange to occur in the future.” If one has a “personal history” with someone else, then process-based trust is generated firsthand. Yet, in many cases, people will rely on secondhand information about past conduct and will refer to “stories, testimonies, evaluations, or credentials given by others” (Sztompka 1999, 71). This is especially true for the initial decision to invest. Investors’ accounts about their motivation to engage with Madoff provide evidence for this kind of trust. Consider, for example, the following statement:

My husband originally knew Mr. Saul Alpern, father-in-law of Mr. Madoff, who was a

very nice honest accountant.… When he mentioned … that his son-in-law was in the

financial area doing very well for his clients, [we] decided to invest.” (Investor 25, impact

statement)

Here, trust is rooted in (a rather vague) knowledge about Madoff’s past performance as an investment manager (i.e., the fact that he was “doing very well for his clients”). Such knowledge may also be obtained from the media, such as when Madoff is praised for his performance and is described as a “wizard”:

As a matter of fact, I went to the New York Public Library and I investigated as best I

could … I do recall there being an article in the New York Times or the Wall Street

Journal, saying the man was a wizard. No one knew exactly how he did what he did,

but he was successful and nobody cared. (Investor 2, interview)

Reputation as a trustworthy investment manager is not necessarily based on a direct assessment of performance, however. It can also derive from someone’s associa- tion with other trustworthy persons or institutions, as the following two quotes exemplify:

We knew [Madoff] had headed up, or started, I think, … the Cincinnati Stock

Exchange. He was a consultant to the SEC. (Investor 9, interview)

Madoff … was also formerly president of NASDAQ and was very highly regarded in

Wall Street. And he [the brother of the investor] couldn’t find out anything negative

about him. (Investor 11, interview)

Madoff was also active in the National Association of Securities Dealers (NASD), a self-regulatory securities industry organization. He served as the Chairman of the Board of Directors and as a member of the Board of Governors of the NASD. He was, further- more, a member of the Board of Directors of the Securities Industry Association, an industry association for securities firms.

Here, it is Madoff’s involvement in prestigious institutions that characterizes him as a trustworthy person. The implicit assumption is that “fame and popularity are achieved

366 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

through exceptional deeds” (Sztompka 1999, 73) and this creates expectations concerning future behavior.

Process-based trust may come about in an even more indirect way. When some of our interviewees referred to the financial expertise of their relatives and friends, they formed their expectations on the basis of prior exchanges with their relatives and friends rather than with Madoff:

First, in about 1985, through a friend of my brother, who … had been very successful

with finances, and my brother was speaking well of him and he had a good fund …,

and I did invest. (Investor 5, interview)

Someone who was “very successful with finances” can be expected to make the right investment decisions in the future. However, recommendations from friends or relatives will not automatically trigger an investment. The level of expectations plays an impor- tant role. Someone who is very trustworthy may recommend an investment with a safe return of, say, 3 percent per annum, but this will hardly motivate an investor who looks for higher returns. In the case of Madoff, it appears that many investors were attracted by the stability of returns that Madoff’s investments had provided in the past rather than looking for an exceptionally high rate of return. As one of our interviewees elaborates:

A lifelong friend of mine, who is one of America’s wealthiest individuals, suggested that

I invest with him [Madoff], because I was looking to put a chunk of money in what I

considered a relatively safe and consistent return, unlike the market where you can go

up 20 percent and down 20 percent …. (Investor 3, interview)

Being “one of America’s wealthiest individuals” implies that this friend would be able to identify a good investment. Process-based trust is at work here. At the same time, the investor was apparently attracted by what he perceived to be “relatively safe and consis- tent returns.” Another investor responds in a similar vein:

Now, my Dad and his brother were very, very, very conservative investors. … Some-

where along the line, probably at one of the country clubs, during a golf game or what-

ever, one of the friends said: “I met this guy Bernie Madoff, and he has a fund that has

virtually guaranteed returns, and he has already got a track record.” (Investor 8,

interview)

It was not extraordinarily high returns that the “conservative investors” were looking for. Rather it was the idea of “virtually guaranteed returns” that were higher than those of other, low-risk investments. Trustworthiness closely relates to consistency and stability of past actions (Sztompka 1999, 72).

How Madoff was constructed as a trustworthy investment manager is also evident from our analysis of the press articles retrieved from Factiva. In the majority of these articles, Madoff is portrayed as a reputed expert in investment management or market making. He features as a pioneer in the field (e.g., the “New York market maker who made the practice [electronic communications networks—ECN] famous” (Stewart 2000)), as the representa- tive of a well-established firm (e.g., his “securities firm is a market maker for many stocks”) (Pearlstein and White 2002) or the chair of an important committee (e.g., Securities Indus- try Association’s trading committee). This reputation of Madoff as an expert is reinforced by the fact that his name is often associated with other well-known players in the financial industry. McTague (1999) recounts that “[Madoff’s] company, which is a market-maker, is

Construction of a Trustworthy Investment 367

CAR Vol. 31 No. 2 (Summer 2014)

building an ECN with Merrill Lynch and Goldman Sachs.” What we see at work here is similar to what Gabbioneta et al. (2013) observe for the Parmalat fraud; namely, that Par- malat apparently managed to conceal its fraudulent activities through its “ability to enhance its status by associating itself with elite third parties.”

As some of the above examples illustrate, the processual dimension of process-based trust may sometimes be implicit, in the sense that the process that grounds trustworthiness is not very visible. Notions such as “expertise” or “reputation” may then easily be regarded as “characteristics” of the person in question. However, since such characteriza- tions are made on the basis of a history of past actions, we follow Zucker (1986) and regard them as instances of process-based trust. This is in contrast to those characteristics that are not really linked to any past exchange or activity but are more intrinsic to a per- son’s life, as discussed next.

Personal ties: Characteristic-based trust

When trust is built on the basis of knowledge about personal characteristics such as one’s family background, age, or ethnicity, Zucker (1986) speaks of it as characteristic-based trust.

18 If another person shares similar characteristics as oneself, then this may create a

common background that reduces the perceived need to negotiate in detail the terms of exchange or to inquire into the other person’s credibility (Zucker 1986, 61). The ascribed characteristics “provide ready-made typifications that, correctly or incorrectly, suggest how individuals with [these] characteristics will behave in certain situations” (Neu 1991a, 187).

Several commentators emphasized how Madoff exploited his affinities with certain groups or communities. It is well-known, for example, that he used his links with the Jewish community to facilitate his fraudulent scheme (Sander 2009, 170; Strober and Strober 2009, 15). Likewise, he had affinities with other groups of significant investors such as motion picture artists, some of them funneled by the Brighton Company, headed by Stanley Chais (e.g., acting couple Kevin Bacon and Kyra Sedgwick, director Steven Spielberg, DreamWorks executive Jeffrey Katzenberg, screenwriter Eric Roth) (Ross 2009, 162, 176); the “French connection” through Access International and Thierry de la Ville- huchet; and the “Latin connection” via Banco Santander (Sander 2009, 83). Fairfax (2001, 70) refers to the exploitation of such shared characteristics as “affinity fraud.” Affinity frauds prey on groups—religious, ethnic, professional, or other like-minded orga- nizations—in order to sell some kind of investment or membership in something (Sander 2009, 73). Among our interviewees and the analyzed statements, only one referred to affinity with Madoff:

My dad did not want to even discuss it. He was so committed to Madoff. And a lot of

this had to do, again, the relationship. … And he had only met Bernie and/or Peter one

time. But the tie to the Jewish community, Jewish philanthropy, Jewish charities and the

fact that his friends were also successfully working with Madoff; there was no other

place for us to put the money. (Investor 8, interview)

Even in cases where there was no direct affinity between an investor and Madoff, characteristic-based trust may have been at work. Potential investors’ trust in their rela- tives or friends, with whom they share certain expectations, can create characteristic-based trust “secondhand.” This seems to have been the case for investor 4 who explains that he trusted his friend’s recommendation:

18. Zucker (1986) uses the notion “characteristic-based trust” while Neu (1991b, 1991a) speaks of “character-

based trust”. We use these notions synonymously.

368 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

I was looking for a new financial advisor, because the one I had had for a few years

was retiring, and I spoke to a businessman, a friend of mine in New York, who told me

that he had two advisors that he used. One was … and another one was Bernard Mad-

off, whom I had never heard of. I had a lot of confidence in my friend. (Investor 4,

interview)

One of the aspects of Madoff’s demeanor that may have allowed him to perpetu- ate a fraudulent scheme for such a long time was that he was active in his community as both a philanthropist and as a contributor and participant to civic organizations. Interestingly, this community engagement did not feature in any of the press articles that we retrieved from Factiva. One explanation for this could be that this type of community engagement is more likely to be mentioned in local or regional media rather than nationwide media. Madoff also served on the board of directors of many charitable institutions (Sander 2009, 55)—several of which entrusted his firm with investing their endowments. He and his wife also gave money to political parties (Oppenheimer 2009, 131), with the major portion going to the Democratic Party of the United States.

Trust in the work of regulators: the SEC

As mentioned above, some of our interviewees consulted publicly available material prior to investing with Madoff in order to verify the trustworthiness of the investment opportu- nity. Their testimonies reveal another source of trust that they relied upon:

There was not a great deal of information available on him [Bernard Madoff], but we

did know that the SEC, the Securities and Exchange Commission here, did give him a

clean bill of health after being investigated. And we decided that was good enough for

us. We are not sophisticated investors. (Investor 2, interview)

He [a friend of the investor] told me that he had talked to the people at the

SEC, and they told him that everything they knew about Mr. Madoff was satis-

factory. (Investor 4, interview)

In the two quotes above, reference is made to the SEC and its role in monitoring the activities of Madoff’s firm. It is apparent from the quotes that the investors were reassured by Madoff’s claim that he had been investigated by the SEC and that he had received “a clean bill of health.” What is at work here is what Zucker (1986) and Neu (1991a) call “institution-based trust.” People rely on the power of institutions—in this case, regulatory bodies—to ensure a proper functioning of social or economic exchanges. The existence of these institutions often produces comfort and reduces the perceived need to monitor or control personally a given exchange process.

As Prentice (2006, 800) explains, “most commentators consider the SEC an extremely successful regulator. The U.S. securities markets are the world’s most efficient and liquid, and inspire a high level of investor confidence. The SEC has received repeated praise throughout its almost seventy-year history as a ‘model agency’.” The trust-producing effects of an institution such as the SEC can partly be inferred also from the way in which the institution portrays itself. On the SEC website (www.sec.gov), one can, for instance, find the following statement:

Crucial to the SEC’s effectiveness … is its enforcement authority. Each year the SEC

brings hundreds of civil enforcement actions against individuals and companies for vio-

lation of the securities laws. Typical infractions include insider trading, accounting

Construction of a Trustworthy Investment 369

CAR Vol. 31 No. 2 (Summer 2014)

fraud, and providing false or misleading information about securities and the companies

that issue them.

Such descriptions may foster investors’ belief that the monitoring by the SEC is com- prehensive and effective. Cases where such monitoring apparently did not work are not mentioned on the website.

Through benefit of hindsight we know that the investors’ trust in the SEC was not warranted. A complete examination of Madoff’s activities was not undertaken by the SEC, and in fact since Madoff’s arrest, the SEC has been criticized for its lack of investi- gative diligence. Concerns about Madoff’s operations began as early as 1999, when a financial analyst named Harry Markopolos told the SEC that he believed it was impossi- ble to achieve the gains Madoff claimed to deliver. Markopolos was ignored by the SEC throughout the 2000 to 2008 time period. He has since published a book about the efforts he made to alert the SEC (Markopolos 2010). The SEC’s Inspector General, H. David Kotz, has revealed that since 1992, the SEC received six substantive complaints raising sig- nificant red flags about Madoff and conducted two investigations and three examinations related to Madoff’s investment advisory business, with none of them leading to discovery of the fraud. In fact, Madoff could have been caught several times, and especially in 2006, but it appears that the investigators never asked the “right” questions (U.S. Securities and Exchange Commission 2009b). As Kotz concludes:

despite numerous credible and detailed complaints, the SEC never properly examined or

investigated Madoff’s trading practices and never took the necessary, but basic, steps to

determine if Madoff was operating a Ponzi scheme. Had these efforts been made with

appropriate follow-up at any time beginning in June of 1992 until December 2008, the

SEC could have uncovered the Ponzi scheme well before Madoff confessed. (U.S. Securi-

ties and Exchange Commission 2009b, 22)

Investors learned about these failures only in retrospect:

You know, I seem to remember that—and I guess that was in the early 90s—there was

some talk of Madoff being investigated by the SEC. But my memory is, my dad checked

it out, and now, we know in retrospect, that if there was an investigation, it took about

ten minutes, and they found no issues, and that was the end of it. (Investor 8, interview)

I continued to feel very secure despite a few blips on the radar, which were immediately

cleared up by a statement made by the SEC confirming that Mr. Madoff was still the

gold standard of Wall Street. (Investor 19, impact statement)

It is difficult to say what exactly prevented the SEC from properly fulfilling its moni- toring role. One of the problems, however, seems to relate to the internal organization of the SEC and its way of handling incoming tips or suspicious facts. These used to come via phone calls, emails, faxes, and even handwritten letters into the SEC’s 11 regional offices and Washington headquarters. Before the Madoff case, one SEC’s office might receive a written complaint about a bad broker, for instance, and put the letter into a filing cabinet if it was deemed without merit. So, if later on a complaint about the same broker was sent to another SEC’s office, staff there would have no easy way of knowing about the earlier tip and connecting the dots. Sometimes, the only way an attorney could find out if someone had looked into a complaint would be to call all the other SEC offi- ces. As a response to its fumbling of early tips about the Madoff fraud, the SEC created the “Tips, Complaints, and Referrals” (or TCR) Database. Once a tip or complaint is

370 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

entered into this database, about 2,300 SEC employees can see it and add new informa- tion. The SEC’s new Office of Market Intelligence, which created a partnership with the Federal Bureau of Investigation in 2011, is using the database as a key tool (Lynch and Goldstein 2011).

Another factor that appears to have played a role in limiting the effectiveness of the SEC in this case relates to its financial resources. As Khuzami (2009) points out, the SEC oversees more than 30,000 registrants, including more than 12,000 public companies, 4,600 mutual fund families, 11,000 investment advisers, 600 transfer agents, and 5,500 broker dealers. In fiscal year 2008, the Enforcement Division received more than 700,000 com- plaints, tips, and referrals regarding potential violations of the federal securities laws. Yet, the entire Enforcement staff nationwide—including lawyers, accountants, information technology staff, and support staff—is just above 1,100 and budget limitations seem to have prevented it from increasing this number. After the Madoff scandal was revealed, the budget of the SEC was initially increased

19 although the House of Representatives subse-

quently pointed out that increasing the SEC’s budget ought to be considered as part of broader efforts to make the agency more effective (Ackerman 2011).

Several of the Madoff investors have filed lawsuits against the SEC (or the United States), arguing that their losses resulted from the insufficient monitoring and control activ- ities of the agency. It is worth mentioning that these lawsuits were not decided in their favor.

20 For example, in the case Dichter-Mad Family Partners, LLP v. U.S.A. and the

SEC, the plaintiffs argued that they had “made their investments in reliance on Madoff’s reputation, clean regulatory record, and the SEC’s implied stamp of approval” (Compl. ¶ 8) and that the SEC’s negligent acts “caused Madoff’s scheme to continue, perpetuate, and expand” (Compl. ¶ 2). The court, however, granted defendants’ motion to dismiss the case. The court confirmed that the allegations against the SEC revealed serious regulatory insufficiencies: “Many of Plaintiffs’ allegations (including the factual averments contained in the Report) identify decisions that, in hindsight, could have and should have been made differently. Other allegations reveal the SEC’s sheer incompetence in regulating Madoff’s broker-dealer, market-making, and investment-management operations” (U.S. District Court Central District of California 2010, 4). However, the court nevertheless dismissed the charges on the basis of the stipulations in the Securities Exchange Act, according to which the SEC has discretion in the timing, scope and manner of how to investigate a given case: “What is lacking in the present Complaint, however, is any plausible allegation revealing that the SEC violated its clear, non-discretionary duties, or otherwise undertook a course of action that is not potentially susceptible to policy analysis” (ibid., 4–5).

In a sense, this type of court decision confirms the dilemma surrounding institution- based trust and controls: while regulatory bodies are expected to fulfill important monitor- ing and control functions, these agencies ultimately cannot be (legally) blamed for not having detected a case of fraud. While investors need to rely on such institution-based mechanisms, they are at the same time expected to not rely blindly on them—and must instead accept responsibility for their own investment decisions.

21

Trust in intermediaries: The role of the auditors

Intermediaries constitute the second institution-trust creating mechanism. They provide market participants with a form of warranty or guarantee that a common set of rules is

19. Http://www.sec.gov/news/press/2009/2009-37.htm, http://www.securitiesdocket.com/2009/07/09/senate-sub-

committee-proposes-to-further-increase-sec-budget-for-2010-to-113-billion (last retrieved: February 1, 2013).

20. To the best of our knowledge. Some cases are still pending as of February 1, 2013.

21. For a more detailed discussion of the principle of ‘individual responsibilization’ in neo-liberal forms of

government, see for instance Miller and Rose (1990), Garland (1996) and O’Malley (1996).

Construction of a Trustworthy Investment 371

CAR Vol. 31 No. 2 (Summer 2014)

being followed and that exchange is possible even in the presence of moral hazard and information asymmetries (Neu 1991b, 249).

Existing accounting research emphasizes the important role of auditors in the produc- tion of trust and comfort (e.g., Pentland 1993; Power 1997; Gendron and B�edard 2006; Malsch and Gendron 2009). By examining financial statements and/or operational activi- ties, auditing adds to the trust of market participants. Yet, the effectiveness of auditing is far from uncontested, given that the quality of an audit ultimately remains difficult to assess. More often than not, it is the very existence of an audit process, rather than any precise knowledge of what is being audited and how, that provides comfort to decision makers. Or, as Power (1997, 64) puts it, “the very idea of audit is valued almost regardless of what is done in its name.” The risky nature of trust placed in auditing is apparent and becomes even more apparent if one considers that, given the complexity of auditing tasks, auditing may easily degenerate into a practice of box-ticking and ritualized verification of superficial facts (Power 1997).

Madoff’s broker-dealer firm (BMIS) was audited by the Certified Public Accountant, David G. Friehling, who was the sole practitioner in the firm Friehling & Horowitz, CPAs, P.C. (F&H). Friehling certified BMIS’s financial statements, including balance sheet, income statement, and cash flow statement, and he wrote a report on BMIS’s inter- nal controls. The financial statements were filed with the SEC and were sent to some of Madoff’s customers who were potentially reassured by the fact that the statements were certified by an auditor (U.S. Securities and Exchange Commission 2009a; see also Henri- ques 2011, 149–50, 254–55). In Appendix 2, we provide an extract of these filings for the year ending October 30, 2007. These include a statement signed by Madoff, the indepen- dent accountant’s report, as well as the statement of financial condition of BMIS.

22

Between 2004 and 2007, Friehling received monthly fees of between $12,000 and $14,500 as compensation for his services.

23 While Friehling was responsible for scrutiniz-

ing the presentation of the financial statements and for identifying any material inadequa- cies in the broker-dealer’s internal control system, it turned out that he did not conduct any independent audits. He simply pretended to have audited Madoff’s operations while, in reality, no such audit was conducted. Accordingly, the SEC charged him in March 2009 with securities fraud, contributing to investment adviser fraud, and the filing of false audit reports with the SEC. Friehling pleaded guilty to all charges in November 2009 but insisted that he did not know about the Ponzi scheme (Henriques 2009b).

24 One of the

factors that arguably allowed the misconduct of Friehling to remain undetected was the lack of information exchange between two regulatory bodies, the SEC and the American Institute of Certified Public Accountants (AICPA). The AICPA requires that auditors undergo a peer review process, including a review of audit work papers. While Friehling (falsely) reported to the SEC as having certified BMIS’s statements, he reported to the AICPA that he did not perform any audits, which allowed him to avoid the peer review.

25

Friehling’s involvement in the Madoff fraud is obvious insofar as he did not conduct any independent audit at all. In comparison, the role of other auditors involved in the case

22. The filings are accessible from the SEC Edgar database (for the year ended October 30, 2007: http://www.

sec.gov/Archives/edgar/data/61369/999999999708001909/9999999997-08-001909-index.htm) (last retrieved:

February 1, 2013).

23. United States District Court Southern District of New York 2009, 2-3. See: http://graphics8.nytimes.com/

images/2009/11/03/business/Friehling.pdf (last retrieved: February 1, 2013).

24. Friehling was to be sentenced for these crimes in February 2010, but because of his cooperation with the

government, his sentence has been postponed several times. At the time of writing this article (February 1,

2013), Friehling had not been sentenced yet.

25. United States District Court Southern District of New York 2009, 6. See: http://graphics8.nytimes.com/

images/2009/11/03/business/Friehling.pdf (last retrieved: February 1, 2013).

372 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

has been more contested. These auditors played an indirect role in that they audited the feeder funds through which investors invested indirectly with BMIS. Among the auditing firms concerned are “leading big-name auditors” (Sarna 2010, 153) such as KPMG, Price- waterhouseCoopers, and Ernst & Young. Several investors have filed lawsuits against these auditors, accusing them of professional malpractice or breach of fiduciary duty. Some commentators agree with these allegations and point to the insufficient controls instituted by the audit firms. Sarna (2010, 153), for example, opines that the auditors of the feeder funds “merely accepted confirmations from Madoff without looking any fur- ther” and cites PricewaterhouseCoopers as an example. Similar criticism has been raised about other auditors (Gandel 2008).

While several investors have sued the auditing firms of the feeder funds, the courts have in most cases granted the defending firms’ requests to dismiss (Gentile 2010; Orrick 2011). A major argument made by the courts is that the investors failed to plead scienter (i.e., to prove that the auditors had an actual intent to deceive, manipulate, or defraud). For instance, in a group of cases concerning feeder funds managed by Tremont Partners, the court established that “a shoddy audit in violation of generally accepted auditing stan- dards (GAAS) does not establish the intent to defraud.”

26 More generally, the court

argued that the auditors were only responsible for auditing the feeder fund and “were never engaged to audit Madoff’s business or to issue an opinion on the financial state- ments of BMIS” (ibid., 13).

The existence of auditors was an additional element in the production of trust in Madoff and his investment firm. On more substantial grounds, however, there was little actual control performed by these auditors. It would appear that investors were comforted by signatures, labels, and “big names,” thereby allowing Madoff to continue with his fraudulent scheme.

Encountering Madoff

The different ways of generating trust that we have examined in the previous sections all share a common element: they apparently contributed to create a sense of trust among investors without active manipulation by Bernard Madoff. Effectively, one could say that Madoff benefited from the malpractice or naivety of other people or agencies that allowed him to perpetuate his fraudulent scheme. Yet, he was certainly more than a passive bystander. As we will argue in this and the following section, Madoff actively provided information that increased his trustworthiness in the eyes of the investors.

The first way in which Madoff influenced investors’ beliefs was through personal encounters with them. In the highly institutionalized setting of today’s economies, personal encounters have lost much of their importance when compared to institution-based forms of trust production. Yet, they are not completely irrelevant. Barrett and Gendron (2006, 636) suggest that institution-based trust “may be sustained or transformed through the way personal level trust is experienced by participants through access points.” On the one hand, face-to-face encounters offer the possibility for a more intimate social relationship in which the contracting parties can subject each other to increased scrutiny (Roberts 1991; Roberts, Sanderson, Barker, and Hendry 2006). On the other hand, they also offer a stage where charisma and persuasive talk can lead to an effective construction of a trustworthy impression.

Most investors did not meet Madoff in person, but some of them did. It is instructive to examine their accounts in terms of how such encounters developed. Among our

26. United States District Court Southern District of New York 2010, 12. See: http://securities.stanford.edu/

1042/TPI_01/2010330_r02x_08CV11212.pdf (last retrieved: February 1, 2013).

Construction of a Trustworthy Investment 373

CAR Vol. 31 No. 2 (Summer 2014)

interviewees, one investor narrated his encounter with Madoff. We quote his narrative at some length here because of its illuminating nature:

I met with Madoff … I met with one of his salespeople, or advisors, I do not remember

the exact word, and I had a meeting with him; and he said to me: “Come on, let us go

meet Bernie now.” So, he took me into Mr. Madoff’s office, and he was sitting behind a

big desk, very imposing, and the first thing he said to me was: “You know, we need a

minimum of a million dollars for you to get into the fund.” And at the time, I had the

money. I said: “I am prepared to do that.” And then, he started asking me a lot of

questions. And I said to him: “It is a very funny thing. I am investing a million dollars

with you and you are asking me the questions. Don’t you think I should be asking you

the questions?” At that point, he said to me: “You do not have to ask me any questions.

We have been interviewed and examined many, many times by the American Securities

and Exchange Commission, and we have got nothing but a clean bill of health from

them, and everything that we do has been supervised and inspected; and we are very,

very regimented, and there is no problem with us. You can just check with the Securities

and Exchange Commission.” (Investor 1, interview)

This quote reflects the different strategies that Madoff used to deflect investors’ efforts to obtain more information. The encounter starts when the potential investor suggests that it should be the investor who asks the questions, rather than Madoff. The first strategy employed by Madoff was to point to the regulatory role played by the SEC. In so doing, he invokes the power of institution-based trust, as discussed above. The beginning of the quote also reveals a complementary mechanism at work. When Madoff starts the conversation by pointing out that one million dollars is required to invest, he creates the image of an exclu- sive investment club—one that is only open to selected individuals. This line of argumenta- tion is continued as the conversation develops, as we see from the rest of the quote:

Then, he said to me: “But I am very sorry. The fund is closed, and we cannot take you

in.” And I said: “Okay, thank you very much.” And I left. I called my accountant and I

said to him: “You know, I met with Mr. Madoff, and the fund is closed; but the funny

thing was, he was asking me the questions and I should have been asking him the ques-

tions.” And he said to me, he just kept referring me back to the Securities and Exchange

Commission. And my accountant said to me: “You know, he says, they have been

inspected many times by the Securities and Exchange Commission and they have a clean

bill of health and the government has found no improprieties and they are a very legiti-

mate company.” The next day afternoon, I received a call from the gentleman whom I

originally met, who took me in to see Mr. Madoff. And he said to me: “You know, I

spoke to Mr. Madoff and you are a very nice fellow, and bla-bla-bla, and we will put

you in the fund. We will get you in.” Like they were doing me a favor. I found out

later, that was their modus operandi with many people. They declined them, and then,

they did them a favor by taking them back in, and I think they have done that many

times. (Investor 1, interview)

The implication that the investor would be allowed into an exclusive club is increased by Madoff first rejecting the request to invest. Only later is Madoff doing the investor a “favor” by eventually allowing him to put money into the fund.

The whole deal with Bernie Madoff was, you felt you are lucky to be in a club, so to

speak, to be invested with him. My friends [said], “Well, can’t you get me in, can’t you

get me in?” (Investor 10, interview)

374 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

The following quote illustrates one strategy of Madoff for not answering questions.

Bernie had several stipulations. He would invest Hadassah’s 27

money but would be

unavailable to answer questions from anyone on our financial advisory board. When I

asked him why, he told me the investment advisory side of his company was very small

and he implied that he was doing this as an accommodation and didn’t want to be both-

ered by people asking him a lot of questions. (Weinstein 2009, 40)

In addition, Madoff’s efforts to limit the information provided to investors were explained on the basis that he had developed an investment strategy that he needed to keep secret in order to continue offering the extra returns that investors were looking for. While none of our interviewees commented on this point, quotes from the secondary liter- ature point in this direction:

The fact that he refused to reveal the secrets of his operation only encourage these inves-

tors to believe that of all the extant investment advisers, he and he alone possessed the

right stuff [quotation from an interview]. (Strober and Strober 2009, 152)

I asked Madoff how he was able to accomplish his amazing returns. “I can’t go into it

in great detail. It’s a proprietary strategy.” (Arvedlund 2009, 7)

By insisting on the proprietary nature of his investment strategy, Madoff managed to become “unaccountable” to his investors regarding the details of his strategy. This is not surprising given that Madoff was regarded as an “expert” or even a “genius” who man- aged to outperform the market precisely because he acted in a different way. Indeed, suc- cessful managers or entrepreneurs are often praised for breaking the rules and acting contrary to conventional wisdom (Messner 2009).

28 If such a practice is accepted—which

is often the case as long as it proves to be successful—then it is difficult to apply the same accountability standards to the details of such actions as in other cases. The designation as an expert or a genius implies that some level of unaccountability must be accepted, because it is understood that experts or geniuses cannot simply “explain” what they do— otherwise, everyone could imitate them.

29

Finally, Madoff was known for his “soft personality,” which he used to reassure inves- tors, as shown in the following quote:

He [Madoff] put his arm around my shoulder and assured me that my money was safe

and I should not worry. I have to admit that I was not sophisticated in investing or

finance and I trusted this kindly man. (Investor 19, impact statement)

27. Hadassah is an American Jewish volunteer women’s organization (charity) involved in health care, educa-

tion and youth programs.

28. As Cohen et al. (2010) note, it appears that several managers of fraudulent firms received praise and admi-

ration from the press.

29. The designation as an “expert” highlights the close association between the three different forms of trust

production, as presented above. One can become an expert through professional certification or one’s asso-

ciation with reputed institutions. In this case, expert status is best characterized as a form of institution-

based trust. In contrast, if one becomes seen as an expert mainly because of one’s past performance, what

is at work is mainly process-based trust. Finally, if expertise is mainly associated with a person’s innate

intelligence or intuition, then it would appear to be a form of characteristic-based trust. All three mecha-

nisms appear to have been at work in the Madoff case, but the role of institutions and the attention given

to past performance were particularly salient, in our view.

Construction of a Trustworthy Investment 375

CAR Vol. 31 No. 2 (Summer 2014)

It therefore appears that the personal encounters between Madoff and some of his investors worked to increase the trustworthiness of the investment opportunity—or at least helped to eliminate concerns that the investors might have had. Madoff successfully mobi- lized his status as an expert (process-based trust), the role of regulatory institutions (insti- tution-based trust), and his appealing personality (characteristic-based trust), to win investors’ confidence.

Account statements

A second way in which Madoff actively influenced investors’ perceptions was through the provision of account statements. While the above evidence suggests that Madoff skillfully avoided providing detailed explanations about his trading strategies, this does not mean that investors were given no information at all. In fact, providing no information would probably have created too much skepticism among investors as to where their money really was going. Madoff’s strategy was to provide investors with a form of written infor- mation that would give them little reason to question the credibility of his operations. From our interviewees, we learn that they received regular statements concerning their investments:

I received a very, very detailed monthly statement, every month without fail. The state-

ment showed purchases of stock. The statement showed selling of stock. The statement

showed purchases of United States Treasury Bills. And some of them even had numbers.

So, you know, you thought you were really getting the right thing. And every month I

looked at my statement, and I said: “Oh, look how much money I made this month.”

You know? And the money just kept coming in on the statements. And because of the

statements, I kept thinking how much money I was making at Madoff’s, which really

did not happen. (Investor 1, interview)

These statements apparently provided comfort to the investors because they showed, in a very detailed way, how much money was invested and earned. This is confirmed by the interviews:

He [Madoff] would show you what you bought and what you sold; or, what he

bought and sold for you; and then, on the final page, basically, a glimpse at a new

balance. (Investor 8, interview)

It is worth pointing out that the importance of the statements as an information source for investors has increased following a change in legislation in 1970 that stipulated that investors would no longer obtain physical delivery of their securities but instead would leave them at their brokerages. Account statements therefore constitute the primary proof of one’s investments.

30

We provide in Appendix 3 excerpts from two consecutive statements. Private informa- tion concerning the investor has been removed.

31 The excerpts allow illustration of how

the investors were informed about their return. Investors received two monthly statements

30. See the testimony of Ron Stein, President of the Network for Investor Action & Protection, dated Septem-

ber 21, 2010, to the Subcommittee on Capital Markets, Insurance and Government Sponsored Enterprises

of the House Financial Services Committee, chaired by Representative Paul E. Kanjorski. Available at:

http://www.investoraction.org/wp-content/uploads/2010/09/Niap_Testimony.pdf (last retrieved: February

1, 2013). See also: http://www.scribd.com/doc/38043731/NIAP-Testimony-Kanjorski-Subcommittee-on-

Capital-Markets-09-23-10 (last retrieved: February 1, 2013).

31. Two investors agreed to send us a copy of their statements: one is dated from 1997, two from 2003 and

the two others from 2008. All are identical in terms of format.

376 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

per account: one securities account and one options account. 32

The monthly beginning and ending balances of each statement are equal but symmetrical.

33 The most important

figure for the investors is the value of the portfolio, which can be found on page 5, under the heading “Market value of securities.” In January 2003, the value of the portfolio of Investor 1 was USD 2,303,088. The following month, the value was USD 2,530,972 (see February 2003, p. 4). The apparent return on the month is thus 5.8 percent. If we consider the ending balance on November 30, 2008 (USD 4,250,725) (not disclosed in Appendix 3), the annual rate of return from January 2003 to November 2008 amounts to 10.5 percent. It seems reasonable to assume that seeing such kind of information provided comfort to investors that their money was invested in a good way. One might argue, of course, that investors should have become suspicious by the continuous positive returns generated and should have questioned whether there are actual transactions behind the statements. We suggest that the lack of such suspicion resulted to an important extent from the appearance of the statements. The statements appeared as if they were real for different reasons.

First, as noted by several of our interviewees, the statements they received were similar or almost identical to those sent by other institutions. There was no apparent difference in Madoff’s statements that would have caused investors to become suspicious:

I received regular information, like I would from any other brokerage firm: monthly

statements, quarterly statements, a year-end statement, the whole thing. Everything

seemed very, very legitimate. … And every time there was supposedly a sale or a pur-

chase, I received a statement accordingly. You know, I had been dealing with brokerage

firms for many years and everything was identical. (Investor 4, interview)

The statements looked just like what you would get from Merrill Lynch, or from Shear-

son, or very similar to what I get today from Fidelity on a retirement account, a small

retirement account. (Investor 8, interview)

A second point worth mentioning is that the statements contained names and stock prices of well-known companies, companies that the investors were familiar with. This ref- erence to real companies reinforced the perceived legitimacy of the statements:

Monthly statements indicated that all of the investments were in the Fortune 500 com-

panies; all very, very, very big names. (Investor 8, interview)

Third, the statements had the appearance of security:

We all knew that there is risk associated with the stock market but our statements

showed [that] we were diversified. (Investor 21, impact statement) 34

32. We have also received examples of these options statements. In Appendix 3, we do not disclose them for

the sake of simplicity and because they are not particularly informative for our purposes.

33. The beginning and ending balance do not represent the value of the portfolio but the balance of a margin

loan account, which is an account with a brokerage firm that allows a customer to buy securities on

credit.

34. Madoff claimed that, at the end of each month, returns were “secured” into U.S. Treasury bills. What

kind of impact this strategy had on the investors is difficult to say, however. On the one hand, one could

imagine that it reinforced investors’ sentiments of security by the fact that the statements where showing

that Madoff parked investors’ money in very secure assets (Treasury bills) at the end of each month. On

the other hand, it may have created some doubt among investors, given that it is a rather peculiar strategy

which involves significant transaction costs without adding really any value to the portfolios.

Construction of a Trustworthy Investment 377

CAR Vol. 31 No. 2 (Summer 2014)

Fourth, the statements sent by BMIS appeared to be real because they could be exchanged with other parties, most notably tax authorities, who would provide another external validation regarding the truthfulness of the statements. Each page of the account statements received by the investors includes the mention “Please retain this statement for income tax purposes” written at the bottom (see Appendix 3).

Regularly, we would get a monthly statement from him. … And once a year, we would

get a summary, a one-page summary of what percentage growth there was, and I used

… those figures … for tax purposes. … The forms were completely consistent and did

not look odd or strange at all. (Investor 5, interview)

At the end of the year, [you got] all the proper tax information that you needed to file

your taxes, because this was a direct investment. (Investor 9, interview)

The production of statements similar to regular bank statements, exchangeable with third parties and linked to other institutions (such as well-known firms) helped to create an impression of normality. Whether the statements were “real” in the sense of represent- ing actual trades was difficult for an investor to see. Some of the investors tried to con- duct “reality checks” by looking at whether the stock prices indicated on the statements corresponded to those in the newspapers or whether the total sums were calculated correctly:

My husband … periodically—especially, when he started to put more money in—…

checked that the stocks on the days that we supposedly purchased them, were purchased

at that price; that the dividends came through on the dates stated. [Madoff] had a very

elaborate scam. (Investor 9, interview)

I used to work in the computer business and I understand bookkeeping; everything was

totaled to the penny. [quotation] (Strober and Strober 2009, 109)

Other investors performed some due diligence:

The monthly statements we received were reviewed and logged in our own version of

due diligence. (Investor 28, impact statement)

While the reality tests performed might have created comfort, they did not allow the investors to judge whether the statements were “true” in the sense of corresponding to an underlying reality of actual trades. Madoff was making sure that there was an ex post cor- respondence between the information on the statements and publicly available informa- tion:

Madoff or his lieutenants were checking the stock returns from previous days and weeks

and instructing the clerks to enter transactions that were based on old results. The com-

puter system would apply the same formula to each client’s account, the only difference

being the number of shares each of them owned. (Kirtzman 2009, 137)

All that investors could check was whether the statements were similar in form to statements issued by other institutions and whether the information regarding stock prices was correct. What they could not do was to “see through” the statements and compare them with some real-world actions to which the statements should correspond. Even when

378 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

supposedly “actual operations” of BMIS were observed, it was apparently impossible to distinguish between appearance and fact

35 :

He [Madoff] impressed me very much, as he did everyone else, and there were banks of

computers and people running all over the place. I thought this had to be legitimate

[quotation]. (Strober and Strober 2009, 131)

What probably added to the perception of trustworthiness was the fact that many investors withdrew money from their accounts. This is likely to have increased their trust in the investment, as it would suggest that the returns existed not only on paper.

36

The above elaborations suggest that the account statements contributed to the crea- tion of process-based trust as they provided information on the performance of the invest- ments. At the same time, their credibility as true representations of reality derived from their association with established institutions, such as the tax administration or other banks that would issue similar statements. There was thus an important element of institu- tion-based trust at work here, the strength of which results from a perceived coherence of information cues. More generally, whether a piece of information is true or not is often not determined by comparing this information to some underlying real event or action. In many cases, such underlying reality is simply not accessible. Instead, representations often achieve credibility when they are consistent with other types of information that are avail- able (Barrett and Gendron 2006). This raises some interesting questions regarding the functioning of financial markets more generally. The discussion that follows takes up this point.

6. Discussion

The construction of trustworthiness

In this paper, we use the investment fraud of Bernard Madoff to inquire into the role of trust in the context of financial markets. As Sztompka (1999, 71) reminds us, the “most common ground for trust is the estimate of the trustworthiness of the target on which we are considering whether to confer trust.” Accordingly, our focus is on examining how the investment opportunity offered by Madoff came to be perceived as trustworthy.

The distinction between process-based, characteristic-based, and institution-based trust (Zucker 1986; Neu 1991b, 1991a) turned out to be helpful for making sense of dif- ferent mechanisms through which trustworthiness can be produced. First, Madoff’s fund had an impressive and long-lasting track record, which apparently fostered investors’ belief that this was an investment with sustainably high performance (process-based trust). An important way in which investors were comforted in this respect was through the regular information provided in the form of account statements. Second, it seems that Madoff’s personal links with some communities and his engagement with social projects helped him gain the image of a trustworthy person (characteristic-based trust). We also found evidence that investors were positively influenced by the way in which Madoff approached them in personal encounters. Third, our empirical evidence suggests that investors were comforted by the existence of regulatory controls and mechanisms of accountability to which Madoff was subject (institution-based trust). In particular, we

35. The Parmalat scandal was also partly based on a system of fictitious documents. Grant Thornton, Parma-

lat’s auditor, has claimed that a letter from Bank of America vouching for €4 billion of cash in Bonlat

was a forgery good enough to fool them (EC 2004).

36. An important number of investors actually withdrew more money than they had originally invested. On

the basis of the so-called “cash-in/cash-out” method, they are designated as “net winners” and are, in con-

trast to the “net losers,” not eligible for any refunds distributed by the Securities Investor Protection Cor-

poration (SIPC).

Construction of a Trustworthy Investment 379

CAR Vol. 31 No. 2 (Summer 2014)

point to the role of the SEC and of auditing firms in confirming Madoff’s trustworthi- ness.

It is difficult to say how strong investors’ trust ultimately was. Evidently, their expecta- tions were strong enough so as to allow them keep their money invested with Madoff—in most cases for a very long time. After all, investment decisions are not one of the events and investments that can, in principle, be undone by withdrawing the money and investing it elsewhere. Hence, there must have been sufficient reason for the investors to maintain their investments with Madoff. Neu (1991a) suggests that expectations tend to be stronger when they build upon information that has been collected firsthand through a personal relationship. We could see the relevance of such intensity at work in the personal encoun- ters between Madoff and his investors. At least as important for the construction of trust- worthiness, however, was the consistency, both in time and space, of the signals obtained by the investors through the more impersonal channels. Consistency in time resulted from the impressive track record that Madoff could claim for his investment and for which he was repeatedly acclaimed in the media. Consistency in space related to the fact that Mad- off and his firm were positively associated with a wide number of reputable institutions, including the SEC, stock exchanges, banks, or industry experts. Such institutional support can strengthen the expectations of the investors regarding the trustworthiness of the investment opportunity (see also Gabbioneta et al. 2013).

The strength of investors’ expectations was thus to an important extent the product of the “collective history” that accumulated over time and that allowed new investors to make an allegedly “informed choice.” The critical role of institutions in creating this col- lective history sheds some important light on previous discussions of the relationship between trust and information. Tomkins (2001), for example, suggests that the creation of trust in a business relationship needs time to develop. Business partners tend not to invest too much in the beginning of their relationship because they do not have much informa- tion about the other party. Over time, each partner accumulates more information about the other, and the level of trust can therefore increase. The existence of institutional mech- anisms that provide secondhand information about the other party influences this relation- ship between trust and information. Since investors can now rely on a history that is not of their own making, they can achieve a high level of trust much more quickly and do not perceive the need to wait and see how the relationship develops.

37 However, this accelera-

tion comes at a risk: the risk that the institutional controls do not work as they are sup- posed to and that the information distributed through institutional channels turns out to be misleading. Such was the case in the Madoff fraud. The representations of Madoff in the media, the reports by regulators and intermediaries, the association of Madoff with reputed third parties—all this created a “reality” into which investors were apparently seduced. They were subjected to an illusion of trustworthiness that institutional mechanisms significantly helped to create.

But is such an illusion symptomatic only for cases of fraud, where investors are appar- ently misled? Or is an element of illusion perhaps inherent to the functioning of financial markets more generally? After all, it would seem that most investment opportunities can- not be assessed independently from the information provided through other market partic- ipants, regulators, the media, or intermediaries. This information is the reality upon which the investors act. And whether these representations “work” or not is not only a question of whether they correspond to some underlying reality but whether a sufficient number of people believe in the representations and, in this sense, “make them work.” So did Madoff

37. This effectively means that the inverse U-shaped curve that Tomkins (2001) suggests for the relationship

between information and trust over time (see his figure 1) is shifted to the left. We thank the Editor for

pointing this out to us.

380 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

perhaps benefit from the fact that financial markets may be, after all, not so different from the functioning of a Ponzi scheme?

Reality and illusion

When investors participate in financial markets, they rely on information that is provided through a variety of channels. While such information is often thought to reduce the need for trust, it is not obvious that it always generates sufficient transparency to allow inves- tors to assess properly the credibility of different investment opportunities or players in the market. In early 2009, the Economist featured an article arguing that the case for more information was not as clear-cut as it may seem:

In financial markets, [transparency] is nearly always equated with information disclo-

sure. The trouble is that the information is often incomplete, irrelevant or outright

incomprehensible. Subprime-mortgage-backed securities are a case in point. These

instruments—whose value remains shrouded in mystery—can have prospectuses of

about 500-600 pages, most of which are devoted to intricate legalese. Yet, inexplicably,

they do not contain the information about individual loans that is needed to detect

default risk. (Anonymous 2009)

Some authors have even gone further by suggesting that financial markets, and the organizations operating in those markets, constitute an arena in which accounting repre- sentations have become the dominant reality upon which people act. Power (2004, 773), for example, suggests that “[o]bjects such as standard costs, returns on investment and net present values move in a virtual world where we no longer remember what any of these things represent, and where increasing numbers of experts are certified to interpret, repre- sent and validate the meaning of these numbers.” They are often still meant to be repre- sentations of a reality, but the existence of this underlying reality is increasingly difficult to ascertain without recurring to just another set of representations or calculations (see also Smith-Lacroix, Durocher, and Gendron 2012). Accordingly, Macintosh et al. (2000) sug- gest the existence of a self-referential system where one (accounting) representation is used to make sense of another one, and where these signs create a form of hyperreality. For instance, when analysts value a firm on the basis of multiples, they use earnings numbers to arrive at market values, but the earnings numbers themselves are (partly) managed by the firms in response to analysts’ expectations. In this way, a self-referential system of cal- culations is created whose values relate to each other, while the reference to the activities that supposedly underlie these representations (production, consumption, etc.) is increas- ingly difficult to pinpoint. The players in the markets may still believe that the representa- tions they act upon are faithful representations of an outside reality. Such belief, however, may eventually turn out to be an illusion.

We need to emphasize at this point that “reality” is not meant to refer to an objective world out there that would be independent of our own representations of it. Social reality —i.e., the set of our collectively established meanings, beliefs, and norms—is not naturally given; rather it is the product of our imagination and continuous reproduction in our actions (Berger and Luckmann 1966; Giddens 1984). If we did not believe in (and act according to) what we take to be our social reality, then this reality would cease to exist. This is true for our legal norms as much as for our economic institutions or for the meanings that we attach to our cultural heritage. In all these cases, we enact reality; that is we create our social world by representing and acting according to it (Weick 1995).

The difference between what is real and what is illusory is therefore not as fundamen- tal as it may appear at first sight. Both are products of our own actions, and both can have real effects in the sense that they influence what we do and how we live. The

Construction of a Trustworthy Investment 381

CAR Vol. 31 No. 2 (Summer 2014)

difference is that illusions have real effects only until they are unveiled as illusions. This happens when these beliefs (or the actions that they motivate) turn out to be incompatible with other parts of our social reality—those parts that we have no reason or possibility to call into question—and, as a consequence, stop being regarded as truthful representations of reality. That is to say, people stop believing in and acting in accordance with these rep- resentations and, in so doing, constitute them as wrong or illusory.

Financial markets illustrate, however, that the line between reality and illusion can be rather blurry. Indeed, we would suggest that the functioning of financial markets relies on our partial indifference regarding the truthfulness of representations (i.e., what is real and what is not). This indifference results from the fact that it is legitimate to exploit represen- tations that later on become regarded as illusory—in fact, it is such exploitation that lies at the very heart of the market mechanism. When an investor sells a stock before everyone else “realizes” that the stock was overpriced compared to the firm fundamentals, the inves- tor has benefited from a reality that was eventually unveiled as an illusion (i.e., a wrong belief that needed to be corrected). Yet, it is rather uncommon to refer to market prices as “real” or “illusory.” It is generally accepted that market prices reflect expectations about the future (Fama 1970) rather than any reality that has already materialized. When expec- tations change, market prices will react. Ex post, expectations that have been corrected may be labeled as “illusory,” but this hardly matters for those who have made or lost money in the market. The illusion had real effects for them.

That an illusion can gain momentum in the first place has something to do with the self-referential nature of financial markets, as alluded to above (see Macintosh et al. 2000). The complexity of financial engineering and the need to rely on information that is difficult or impossible to verify contribute to a temporary decoupling of financial markets from the social reality outside these markets (i.e., expectations and beliefs that are not formed on the same market). We say “temporary,” because at some point, these two reali- ties must be realigned with each other. This happens when market participants adjust their expectations and, in so doing, enact a different reality based on some other information that they have obtained. We can observe such a dynamic when investors create speculative bubbles or contribute to the development of a financial crisis. Indeed, it is worthwhile to compare such cases to what we could observe in the Madoff fraud.

Roberts and Jones (2009) examine the recent subprime crisis and suggest seeing the root of the crisis in the way in which different representations of reality (i.e., accounting models and calculations) were used to create new representations, up to a point where the connections between these derivative representations and some underlying assets were no longer visible. When market participants started to have doubts about the “true” value of their securities, their actions triggered a spiral of distrust. The accounting signs, which hitherto had served as the reality for the investors, “were now viewed as possibly distorting ‘mediators’ ” (ibid., 863) of the “true” reality that was now brought into play. In a moment of panic, investors anticipated that they would incur losses and started to try to get rid of their securities, in an act of “self defense” that turned out to be self-fulfilling.

In a sense, one can see the rise and fall of Madoff’s investment firm in the same light. Luring investors into the fund was possible for Madoff because he had managed to con- struct his investment opportunity as a trustworthy one—but more fundamentally because investors were used to accepting the world of representations and expectations as their dominant reality. Despite warning signals issued by individuals such as Markopolos (2010), the missing link between the representations on the financial market and the reality outside these markets remained undetected in the Madoff fraud. The available information pointed toward the same direction (i.e., that Madoff was a trustworthy investment man- ager). The investment opportunity offered by Madoff became a “phantastic object”

382 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

(Eshraghi and Taffler 2012) in the sense that the fascination with its past, present, and promised future contributed to prevent detection of the fraud.

That some investors withdrew their money, and in so doing successfully returned to the “real world,” was not a problem for Madoff as long as most others did not ask to withdraw their money and as long as he could attract a sufficient number of new investors. The Ponzi scheme collapsed only when, in the middle of the financial crisis, many of his investors wanted to withdraw their funds. In early December 2008, Madoff informed a senior employee of his firm that there had been requests from investors to withdraw funds in the amount of approximately $7 billion. Madoff was struggling to obtain the liquidity to meet these obligations and eventually confessed to his sons that he had “absolutely nothing” and that it was “just one big lie.”

38 For Madoff and his investors, this was the

painful return to “reality.” Investors’ expectations, which had hitherto constituted the real- ity for them, were now brutally downgraded so as to align them with what turned out to be the stronger reality (i.e., that Madoff could not deliver on his promises because he operated a fraudulent scheme). The difference with respect to the case of speculative bub- bles and financial crises is one of size rather than substance. In the case of Madoff, there never was an actual link between the representations he created and the reality that these were supposed to represent in the eyes of the investors. In the case of speculative bubbles, such a relationship exists, but it becomes strained over time. The speculative bubble bursts when investors start to get out of their investments and, in so doing, send important signals to other market participants, who then follow. The subprime crisis that was built upon “a mutually self-referencing hall of mirrors” materialized only when some people were no longer willing to accept the representations as good representations of reality and then “real defaults … started to occur” (Roberts and Jones 2009, 864). Madoff’s Ponzi scheme, in turn, collapsed when investors withdrew their money and Madoff became illiquid.

Hence, one could perhaps conclude that Madoff’s investors did nothing other than what most of us do when we participate in financial markets: to participate in the creation of a reality that is based on expectations. For some of the investors with Madoff (those who withdrew all their money before the fraud was unveiled), there was no real difference from other investments—they invested their money at one point and withdrew it at another. If we did not have traces of Madoff’s activities (or nonactivities) in the stock market, we would never know, in principle, whether he had really invested money for these clients or not. Per- haps, some of his investment activities were actually “real” in the sense of not being based on a Ponzi scheme. Or perhaps there were only fake activities. It is remarkable that this does not, in principle, really matter to the “lucky” investors. Reality and illusion effectively become indistinguishable at this point. Their reality is a hyperreality (McGoun 1997).

39

We would thus suggest seeing the Madoff fraud as an extreme case (Flyvbjerg 2001) that illustrates, in a dramatic way, an element of illusion that, we believe, characterizes the functioning of financial markets more generally. The construction of trustworthiness is the product of our own making, and so is the return to a different reality once the constructed one is unveiled as an illusion.

This does not mean, to be sure, that one should be indifferent to such cases of fraud and just treat them as an inevitable by-product of the functioning of financial markets.

38. United States District Court Southern District of New York against Bernard Madoff, BMIS (http://www.

sec.gov/litigation/complaints/2008/comp-madoff121108.pdf) (last retrieved: February 1, 2013).

39. Note, however, that in the case of fraud, a different reality may be introduced ex post, with important

consequences for investors. This happened in the Madoff case when the trustee defined the clawback mech-

anism that implied that investors with a positive net cash flow (the “net winners”) were required to pay

back this net cash flow. In other words, the formerly “lucky” investors who had gained money on the

basis of an illusion which they took for reality were now confronted with a new reality that uncovered this

illusion.

Construction of a Trustworthy Investment 383

CAR Vol. 31 No. 2 (Summer 2014)

We can still reflect upon possible ways of how to try to mitigate the painful consequences that such cases obviously imply for investors and society. The following section is dedi- cated to offering some reflections in this respect.

Practical implications

In light of the dramatic implications of the Madoff fraud, it would appear natural to ask how such a case can be avoided in the future. The first, and somewhat sobering, answer would probably have to be that such cases of fraud can never be completely avoided. For as soon as the participation in an exchange requires trust, there is always the possibility that such trust will be abused. Without the possibility of deceit, trust would not be necessary in the first place.

Yet, this is hardly an answer that regulators and representatives of state and market agencies can accept. As the Madoff case demonstrates, an important part of the trustwor- thiness that investors rely on is produced through actors such as the SEC or auditors, whose role is to monitor and control market participants. The more often they fail in properly fulfilling this role, the more damage will be done to investors, and the more the marketplace will suffer in terms of credibility and trustworthiness (Unerman and O’Dwyer 2004). This does not necessarily mean that investors would avoid financial markets and invest their money in real assets instead. However, it does mean that it will become more difficult for them to distinguish between more and less credible investment opportunities, as the available “system of distinctions” can no longer be trusted. While this would nega- tively affect the functioning of financial markets in general, one may also see some positive element in it. For if institutions lose some of their credibility as trust-producing mecha- nisms, investors would apparently have to accept more responsibility for themselves. In a sense, there is a paradox at work here. Good institutional controls are necessary for the functioning of the financial markets overall. But the better the institutional controls work overall, the more investors have reason to trust them, which may actually result in more dramatic losses in those cases where the controls do not work. This is because investors may too easily accept institutionally-produced information as credible, which would lower their perceptions of risk when considering a particular investment opportunity.

Perhaps the most appropriate way to address the problem of fraud is then to work on two fronts in parallel. On the one hand, it seems critical to have well-functioning regula- tory bodies that are equipped with sufficient resources to carry out detailed monitoring and due diligence. The point here is not to promise more comprehensive controls but to deliver on what is already promised. In particular, it would seem important to ensure that controls effectuated by different regulatory bodies are indeed independent from each other, in order to avoid that control deficiencies are transmitted from one agency to the other. In a similar way, the concerns and doubts of third parties should be taken seriously.

On the other hand, there is a need to be aware that even the best controls will some- times fail. This insight may motivate actions at the institutional and individual level that are concerned with limiting the consequences that such (rare) cases would imply. One reaction on the institutional level could be to restrict the size of the portfolio that investment advisors are allowed to manage. While this would probably reduce the efficiency of investment funds, it would at the same time keep the potential damage small. This argument would apply to cases of fraud in the same way as it does to cases of financial distress, where financial organizations that are considered “too big to fail” create serious problems for the state.

On the individual level, the best way to limit one’s potential losses is perhaps still the well-established (but often not adhered to) rule of diversification (see Makower and Marschak 1938; Markowitz 1952; Samuelson 1967). Diversification allows investors to reduce risk by holding a variety of different assets in their portfolio. The diversification principle can also be applied to the intermediaries used by investors. Even if investors can- not entirely rule out cases of fraud, diversification among different intermediaries at least

384 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

allows them to reduce the relative impact of such events on their wealth. Another possibil- ity would be explicitly to foreclose particular forms of exchange. Our interview with the CFO of an organization that decided not to invest with Madoff provides an example of such a mechanism. The CFO explained that his organization never considered investing with Madoff because they “had a firm policy against investing in funds that were managed by members [of their investment committee]” and because they required an independent custodian for any management account investment (Noninvestor 1). Rules of this nature prevent against entering into certain types of investments even when there are “good” rea- sons to do so, such as high past performance, reputation, or an exclusive right to invest.

7. Conclusion

Recent research on financial markets suggests that the level of trust correlates with the extent of stock market participation, both at the individual and the country level (Guiso et al. 2008). Moreover, trust is found to be a function of individual characteristics (such as education or income) as well as characteristics of the community from which one comes (Alesina and La Ferrara 2002). Our paper contributes to this literature by examining the production of trust. Rather than looking at the characteristics of those who are trusting (the “subjects”), we consider the mechanisms through which investment opportunities (the “objects”) are constructed as trustworthy ones.

Our findings suggest that the production of trustworthiness is the outcome of different mechanisms at the institutional as well as individual level. In particular, we point to the con- sistency of information cues in time and space that apparently comforted investors to put their money into Madoff’s fund. The Madoff case illuminates how the provision of informa- tion can lead to an “illusion of trustworthiness” that is difficult for investors to escape. An element of such illusion, we suggest, is inherent to the functioning of financial markets more generally, which might explain part of the difficulty of detecting a financial scam.

There are different ways in which the findings from the current research may be com- plemented or refined in future studies. One possibility would be to consider in more detail those who trust. In particular, a more comprehensive consideration of the “life projects” of investors (involving, for example, their experiences with investments in the past) could shed additional information on their decisions on whether and how to invest. We can only encourage future research to explore this possibility and follow people’s investment deci- sions over a longer period of time. Such longitudinal studies have provided rich insights in sociology and policy analysis (e.g., Halsey, Heath, and Ridge 1980), and they may prove fruitful for our understanding of people’s behavior in financial markets.

Another direction for future research could be to examine in more detail the role of intermediaries, such as banks or investment advisors, in the production of trustworthiness. In the Madoff case, a large number of investors were indirectly invested in the Madoff fund and were arguably in some way guided by their financial advisers. It is unclear to what extent such intermediaries actually perform due diligence when recommending invest- ment opportunities and to what extent their advice influences investors’ decisions.

It may also be worthwhile to investigate in more detail how trust in institutions is maintained in spite of failures experienced in the functioning of such institutions. Moore et al. (2006), for example, propose a political psychological framework, the “issue-cycle theory,” according to which organizations recover from public outrage after having obtained undue advantages following a strong lobbying. Empirically, Humphrey et al. (1992) analyze the strategies implemented by the British audit profession in response to the audit expectation gap: one defensive, focusing on education and reassurance of the public; the other one being “constructive” based on the change of audit activities. In our view, there is more scope to examine how organizations such as regulators, auditors, or the media manage to maintain or reestablish their reputations as trustworthy institutions.

Construction of a Trustworthy Investment 385

CAR Vol. 31 No. 2 (Summer 2014)

The functioning of financial markets encompasses much more than the alignment of demand and supply. In this paper, we have attempted to improve our understanding of the social construction of trustworthiness that underlies investors’ decisions in such markets. We can only encourage further research that engages with financial markets in a similar way.

Appendix 1

Interview guide

Before the interview starts: - Explain the purpose of the interview: to gain a better understanding of what happened between Madoff and investors; stress the academic purpose.

- Explain that we are interviewing several investors and will treat them all anonymously. Names, investment sums, etc., won’t be disclosed.

- Ask for permission to record the interview, as this is important for the academic purpose.

Questions: 1. To start with, could you please explain how you were affected by Madoff’s fraud? - When did you first invest with Madoff? - How did you invest? Directly or indirectly? Through whom? - Why did you invest? - Do you know other people who had invested with Madoff? Whom?

2. Why did you invest money with Madoff? - Did you know that you had invested money with him? - Had you heard about Madoff before? From the media? From other people? - Did you know anything about the performance of Madoff’s investment funds before you invested? If yes, what did you think about this performance?

3. How was the performance of your investment (before the fraud was detected)? - What did you think about this performance?

4. Did you ever have doubts about your investment? - If yes, why? And what did you do in response to these doubts? - If no, why not?

5. What kind of information did you receive from Madoff regarding your investment? - Records/reports? If yes,

• What was stated on these records? • What did you think about these records?

6. Did you ever meet Madoff personally or talk to him? (See also question 8.) - If yes, what did you talk about? - What impression did he make on you? - Did the conversation increase or decrease your trust in the investment? 7. Did you meet some of his employees or talk to them? - If yes, what did you talk about?

8. Did you ever ask Madoff or one of his employees for more information regarding your investment?

- If yes: What kind of information did you ask for? ? Ask interviewee to narrate the episode in detail.

• What was the answer? • Did the answer satisfy you? At which point did your inquiry stop? Why?

- If no: Why not?

386 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

Appendix 2

Madoff filings for the year 2007 with the auditor’s report

Construction of a Trustworthy Investment 387

CAR Vol. 31 No. 2 (Summer 2014)

388 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

Construction of a Trustworthy Investment 389

CAR Vol. 31 No. 2 (Summer 2014)

Appendix 3

Example of individual investor’s account statement

Part 1 - January 2003

390 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

Part 2 - February 2003

Construction of a Trustworthy Investment 391

CAR Vol. 31 No. 2 (Summer 2014)

392 Contemporary Accounting Research

CAR Vol. 31 No. 2 (Summer 2014)

References

Ackerman, A. 2011. House panel’s head backs budget increase for SEC. Wall Street Journal., Sep-

tember 15. Available at: http://online.wsj.com/article/SB100014240531119040606045765725833

17254882.html.

Ahrens, T., and C. S. Chapman. 2006. Doing qualitative field research in management accounting:

Positioning data to contribute to theory. Accounting, Organizations and Society 31 (8): 819–41.

Alesina, A., and E. La Ferrara. 2002. Who trusts others? Journal of Public Economics 85 (2):

207–34.

Alvesson, M. 2003. Beyond neopositivists, romantics, and localists: A reflexive approach to inter-

views in organizational research. Academy of Management Review 28 (1): 13–33.

Anonymous. 2009. Full disclosure—the case for transparency in financial markets is not clear-cut.

The Economist 390 (8619): 78.

Appelbaum, B., D. S. Hilzenrath, and A. R. Paley. 2008. All just one big lie. Washington Post,

December 12. Available at http://www.washingtonpost.com/wp-dyn/content/article/2008/12/12/

AR2008121203970.html?hpid=topnews.

Arrow, K. J. 1963. Uncertainty and the welfare economics of medical care. American Economic

Review 53 (5): 941–73.

Arrow, K. J. 1984. The economics of information. Cambridge: Harvard University Press.

Arvedlund, E. 2001. Don’t ask, don’t tell: Bernie Madoff is so secretive, he even asks his investors to

keep mum. Barron’s, May 7, 26–7.

Arvedlund, E. 2009. Madoff—the man who stole $65 billion. London: Penguin Books.

Arvedlund, E., ed. 2010. The club no one wanted to join—Madoff victims in their own words.

Andover, MA: The Doukathsan Press of the Massachusetts School of Law.

Ball, R., and P. Brown. 1968. An empirical evaluation of accounting income numbers. Journal of

Accounting Research 6 (2): 159–78.

Barrett, M., and Y. Gendron. 2006. Webtrust and the “commercialistic auditor”: The unrealized

vision of developing auditor trustworthiness in cyberspace. Accounting, Auditing & Accountabil-

ity Journal 19 (5): 631–62.

Berger, P. L., and T. Luckmann. 1966. The social construction of reality: A treatise in the sociology

of knowledge. Garden City, NY: Anchor Books.

Bernard, C., and P. Boyle. 2009. Mr. Madoff’s amazing returns: An analysis of the split-strike

conversion strategy. Journal of Derivatives 17 (1): 62–76.

Bohnet, I., B. Herrmann, and R. Zeckhauser. 2010. Trust and the reference points for trustworthi-

ness in Gulf and Western countries. Quarterly Journal of Economics 125 (2): 811–28.

Callon, M. 2009. Civilizing markets: Carbon trading between in vitro and in vivo experiments.

Accounting, Organizations and Society 34 (3–4): 535–48.

Carlin, B. I., F. Dorobantu, and S. Viswanathan. 2009. Public trust, the law, and financial invest-

ment. Journal of Financial Economics 92 (3): 321–41.

Cohen, J., Y. Ding, C. Lesage, and H. Stolowy. 2010. Corporate fraud and managers’ behavior:

Evidence from the press. Journal of Business Ethics 95 (Supplement 2): 271–315.

Cooper, D. J., and W. Morgan. 2008. Case study research in accounting. Accounting Horizons 22

(2): 159–78.

EC. 2004. Milking lessons—the crisis at Parmalat. The Economist, January 3, 45–46.

Ericsson, K. A., and H. A. Simon. 1980. Verbal reports as data. Psychological Review 87 (3):

215–51.

Eshraghi, A., and R. Taffler. 2012. Hedge funds and unconscious fantasy. Accounting, Auditing &

Accountability Journal 25 (8): 1244–65.

Fairfax, L. M. 2001. “With friends like these”: Toward a more efficacious response to affinity-based

securities and investment fraud. Georgia Law Review 36: 63–119.

CAR Vol. 31 No. 2 (Summer 2014)

Construction of a Trustworthy Investment 393

Fama, E. 1970. Efficient capital markets: A review of theory and empirical work. Journal of Finance

25 (2): 383–417.

Flyvbjerg, B. 2001. Making social science matter. Cambridge: Cambridge University Press.

Frank, R., A. Efrati, A. Lucchetti, and C. Bray. 2009. Madoff jailed after admitting epic scam. Wall

Street Journal, March 13: Available at http://online.wsj.com/article/SB1236856934499065

51.html?mod=djemalertNEWS.

Gabbioneta, C., R. Greenwood, P. Mazzola, and M. Minoja. 2013. The influence of the institutional

context on corporate illegality. Accounting, Organizations and Society. 38 (6–7): 484–504.

Gandel, S. 2008. The Madoff fraud: How culpable were the auditors? Time, December 17. Available

at: http://www.time.com/time/business/article/0,8599,1867092,00.html.

Garland, D. 1996. The limits of the sovereign state: Strategies of crime control in contemporary soci-

ety. British Journal of Criminology 36 (4): 445–71.

Gendron, Y., and J. B�edard. 2006. On the constitution of audit committee effectiveness. Accounting,

Organizations and Society 31 (3): 211–39.

Gendron, Y., and L. F. Spira. 2009. What went wrong? The downfall of Arthur Andersen and the

construction of controllability boundaries surrounding financial auditing. Contemporary

Accounting Research 26 (4): 987–1027.

Gentile, V. 2010. Federal court dismisses investors’ claims against auditors of Madoff feeder funds.

DrinkerBiddle, July 29. Available at: http://www.drinkerbiddle.com/resources/publications/2010/

legal-briefs?Page=4&Section=Publications&Year=&Practice=0&Attorney=0.

Giddens, A. 1984. The constitution of society: Outline of the theory of structuration. Cambridge:

Polity Press.

Giddens, A. 1990. The consequences of modernity. Stanford, CA: Stanford University Press.

Golden-Biddle, K., and K. Locke. 2007. Composing qualitative research (2nd ed.). Thousand Oaks,

CA: Sage.

Guiso, L., P. Sapienza, and L. Zingales. 2008. Trusting the stock market. Journal of Finance 63 (6):

2557–600.

Halsey, A. H., A. F. Heath, and J. M. Ridge. 1980. Origins and destinations: Family class and

education in modern Britain. Oxford: Clarendon Press.

Heider, F. 1958. The psychology of interpersonal relations. New York: John Wiley & Sons.

Henriques, D. B. 2009a. Madoff sentenced to 150 years for Ponzi scheme. New York Times, June 29.

Available at http://www.nytimes.com/2009/06/30/business/30madoff.html?_r = 1&hp.

Henriques, D. B. 2009b. Madoff’s accountant pleads guilty in scheme. New York Times, November 4.

Available at: http://www.nytimes.com/2009/11/04/business/04madoff.html?ref=davidgfriehling.

Henriques, D. B. 2011. The wizard of lies—Bernie Madoff and the death of trust. New York: Times

Books–Henry Holt and Company.

Huber, G. P., and D. J. Power. 1985. Retrospective reports of strategic-level managers: Guidelines

for increasing their accuracy. Strategic Management Journal 6 (2): 171–80.

Humphrey, C., P. Moizer, and S. Turley. 1992. The audit expectations gap—plus ca change, plus

c’est la meme chose? Critical Perspectives on Accounting 3 (2): 137–61.

Khuzami, R. 2009. Testimony concerning strengthening the SEC’s vital enforcement responsibilities by

director, division of enforcement, U.S. Securities and Exchange Commission before the U.S. Sen-

ate banking, housing, and urban affairs subcommittee on securities, insurance, and investment.

May 7. Available at: http://www.sec.gov/news/testimony/2009/ts050709rsk.htm.

Kirtzman, A. 2009. Betrayal—the life and lies of Bernie Madoff. New York: HarperCollins.

Latour, B. 1987. Science in action. Milton Keynes: Open University Press.

LeBor, A. 2009. The believers—how America fell for Bernard Madoff’s $65 billion investment scam.

London: Weidenfeld & Nicolson.

Lieberman, D., P. Gogoi, T. Howard, K. McCoy, and M. Krantz. 2008. Investors remain amazed

over Madoff’s sudden downfall. USA Today, December 15. Available at: http://www.usatoday.

com/money/markets/2008-12-14-ponzi-madoff-downfall_N.htm.

CAR Vol. 31 No. 2 (Summer 2014)

394 Contemporary Accounting Research

Lincoln, Y. S., and E. G. Guba. 1985. Naturalistic inquiry. Newbury Park, CA: Sage Publications.

Lukka, K., and S. Modell. 2010. Validation in interpretive management accounting research.

Accounting, Organizations and Society 35 (4): 462–77.

Lynch, S. N., and M. Goldstein. 2011. SEC modernizing handling of tips, complaints after Madoff

mistakes. July 27. Available at: http://www.huffingtonpost.com/2011/07/27/sec-learns-lesson-

from-madoff-tip-division_n_910630.html.

Macintosh, N. B., T. Shearer, D. B. Thornton, and M. Welker. 2000. Accounting as simulacrum

and hyperreality: Perspectives on income and capital. Accounting, Organizations and Society 25

(1): 13–50.

MacKenzie, D. 2009. Making things the same: Gases, emission rights and the politics of carbon

markets. Accounting, Organizations and Society 34 (3–4): 440–55.

Makower, H., and J. Marschak. 1938. Assets, prices and monetary theory. Economica 5 (19):

261–88.

Malsch, B., and Y. Gendron. 2009. Mythical representations of trust in auditors and the preserva-

tion of social order in the financial community. Critical Perspectives on Accounting 20 (6): 735–

50.

Markopolos, H. 2010. No one would listen: A true financial thriller. Hoboken, NJ: John Wiley &

Sons.

Markowitz, H. 1952. Portfolio selection The Journal of Finance 7 (1): 77–91.

McCool, G., and E. Honan. 2009. Madoff to appeal bail, net worth revealed. Reuters, March 13.

Available at http://www.reuters.com/article/idUSN1343149520090313.

McGoun, E. G. 1997. Hyperreal finance. Critical Perspectives on Accounting 8 (1–2): 97–122.

McTague, J. 1999. A stockholder-owned big board: Too famous to fail? Banking reform and life

insurers. Barron’s, November 1, B 34.

Messner, M. 2009. The limits of accountability. Accounting, Organizations and Society 34 (8):

918–38.

Meyer, J. W., and B. Rowan. 1977. Institutionalized organizations: Formal structure as myth and

ceremony. The American Journal of Sociology 83 (2): 340–63.

Miller, P., and N. Rose. 1990. Governing economic life. Economy and Society 19 (1): 131–62.

Moore, D. A., P. E. Tetlock, L. Tanlu, and M. H. Bazerman. 2006. Conflicts of interest and the case

of auditor independence: Moral seduction and strategic issue cycling. Academy of Management

Review 31 (1): 10–29.

Neu, D. 1991a. New stock issues and the institutional production of trust. Accounting, Organizations

and Society 16 (2): 185–200.

Neu, D. 1991b. Trust, contracting and the prospectus process. Accounting, Organizations and Society

16 (3): 243–56.

O’Malley, P. 1996. Risk and responsibility. In Foucault and political reason: Liberalism, neo-liberal-

ism and rationalities of government, eds. A. Barry, T. Osborne and N. Rose, 189–207. London:

UCL Press.

Ocrant, M. 2001. Madoff tops charts; skeptics ask how. MAR/Hedge 89 (May): 1–5.

Olsen, R. A. 2008. Trust as risk and the foundation of investment value. Journal of Socio-Economics

37 (6): 2189–200.

Oppenheimer, J. 2009. Madoff with the money. Hoboken, NJ: John Wiley & Sons.

Orrick. 2011. Securities class action against outside auditor of a Madoff feeder fund dismissed for

failure to plead scienter. Weekly Auditor Liability Bulletin, February 4. Available at: www.or-

rick.com/Events-and-Publications/Documents/3967.pdf.

Pearlstein, S., and B. White. 2002. Dow rises 489 as stocks surge. Washington Post, July 25, A01.

Pentland, B. T. 1993. Getting comfortable with the numbers: Auditing and the micro-production of

macro-order. Accounting, Organizations and Society 18 (7–8): 605–20.

Power, M. 1997. The audit society: rituals of verification. Oxford: Oxford University Press.

CAR Vol. 31 No. 2 (Summer 2014)

Construction of a Trustworthy Investment 395

Power, M. 2004. Counting, control and calculation: Reflections on measuring and management.

Human Relations 57 (6): 765–83.

Pratt, M. G. 2009. For the lack of a boilerplate: Tips on writing up (and reviewing) qualitative

research. Academy of Management Journal 52 (5): 856–62.

Prentice, R. A. 2006. The inevitability of a strong SEC. Cornell Law Review 91 (4): 775–839.

Roberts, J. 1991. The possibilities of accountability. Accounting, Organizations and Society 16 (4):

355–68.

Roberts, J., and M. Jones. 2009. Accounting for self interest in the credit crisis. Accounting,

Organizations and Society 34 (6–7): 856–67.

Roberts, J., P. Sanderson, R. Barker, and J. Hendry. 2006. In the mirror of the market: The disci-

plinary effects of company/fund manager meetings. Accounting, Organizations and Society 31

(3): 277–94.

Ross, B. 2009. The Madoff chronicles—inside the secret world of Bernie and Ruth. New York:

Hyperion.

Safer, M. 2009. The Madoff scam: Meet the liquidator. 60 Minutes (CBS News): Available at http://

www.cbsnews.com/stories/2009/09/24/60 minutes/main5339719.shtml?tag=currentVideoInfo;

segmentUtilities.

Samuelson, P. A. 1967. General proof that diversification pays. Journal of Financial & Quantitative

Analysis 2 (1): 1–13.

Sander, P. 2009. Madoff—corruption, deceit, and the making of the world’s most notorious Ponzi

scheme. Guilford: The Lyons Press.

Sapienza, P., and L. Zingales. 2012. A trust crisis. International Review of Finance 12 (2): 123–31.

Sarna, D. E. Y. 2010. History of greed—financial fraud from Tulip Mania to Bernie Madoff.

Hoboken, NJ: John Wiley & Sons.

Seal, M. 2009. Madoff’s world. Vanity Fair, April. Available at http://www.vanityfair.com/politics/

features/2009/04/madoff200904.

Silverman, D. ed. 2004. Qualitative research: Theory, method and practice. London: Sage Publica-

tions.

Smith-Lacroix, J.-H., S. Durocher, and Y. Gendron. 2012. The erosion of jurisdiction: Auditing in a

market value accounting regime. Critical Perspectives on Accounting 23 (1): 36–53.

Stewart, J. K. 2000. Brokers order flow choice may inflate share price. Knight-Ridder Tribune

Business News, August 26.

Strober, D., and G. Strober. 2009. Catastrophe—the story of Bernard L. Madoff, the man who

swindled the world. Beverly Hills: Phoenix Books.

Sztompka, P. 1999. Trust: A sociological theory. Cambridge: Cambridge University Press.

Tomkins, C. 2001. Interdependencies, trust and information in relationships, alliances and networks.

Accounting, Organizations and Society 26 (2): 161–91.

U.S. District Court Central District of California. 2010. Dichter-Mad family partners, LLP v. U.S.A.

and the SEC. Available at: http://www.jenner.com/system/assets/assets/2605/original/Dich-

ter_Mad_Family.pdf?1319148148.

U.S. Securities and Exchange Commission. 2009a. Litigation release No. 20959—Accounting and

Auditing Enforcement Release No. 2992, March 18. Available at http://www.sec.gov/litigation/li-

treleases/2009/lr20959.htm.

U.S. Securities and Exchange Commission. 2009b. Report of investigation executive summary.

Available at http://www.sec.gov/spotlight/secpostmadoffreforms/oig-509-exec-summary.pdf.

U.S. Securities and Exchange Commission. 2010. Ponzi schemes. Available at http://www.sec.gov/

answers/ponzi.htm.

Unerman, J., and B. O’Dwyer. 2004. Enron, WorldCom, Andersen et al.: A challenge to modernity.

Critical Perspectives on Accounting 15 (6/7): 971–93.

Vollmer, H., A. Mennicken, and A. Preda. 2009. Tracking the numbers: Across accounting and

finance, organizations and markets. Accounting, Organizations and Society 34 (5): 619–37.

CAR Vol. 31 No. 2 (Summer 2014)

396 Contemporary Accounting Research

Voreacos, D., and D. Glovin. 2008. Madoff confessed $50 billion fraud before FBI arrest. Bloom-

berg, December 12. Available at http://www.bloomberg.com/apps/news?pid=newsar-

chive&sid=atUk.QnXAvZY.

Weick, K. E. 1995. Sensemaking in organizations. Thousand Oaks, CA: Sage Publications.

Weinstein, S. 2009. Madoff’s other secret—love, money, Bernie and me. New York: St Martin’s Press.

Wilcox, R. T. 2003. Bargain hunting or star gazing? Investors’ preferences for stock mutual funds.

Journal of Business 76 (4): 645–63.

Zingales, L. 2009. The future of securities regulation. Journal of Accounting Research 47 (2):

391–425.

Zucker, L. G. 1986. Production of trust: Institutional sources of economic structure, 1840–920.

Research in Organizational Behavior 8: 53–111.

CAR Vol. 31 No. 2 (Summer 2014)

Construction of a Trustworthy Investment 397