Conflicts of Interest in Business: A Review of the Concept

All companies admit in their codes of conduct that conflicts of interest (CIs) are a threat to their efficiency, integrity and reputation. Except for insider trading, definitions of CIs are strictly particular to each business and publicly expressed through their codes of ethics. I propose an interpretative analysis of what is understood by conflict of interest in the codes of ethics of the world’s largest companies, along with a comprehensive review of CIs in several sections: employment, contracting, corporate assets, insider trading and personal investments, competitors, and corporate image. The present paper offers solutions to avoid or resolve CIs in a business context, by combining economic preference with the psychological cognitivist view of self-interest. The conclusion is that a code of ethics and relevant training are protective measures for a company wishing to convince its employees that they are better off not entering CIs.


Introduction
Labelling a situation a "conflict of interest" (CI) is part of common speech, and it is also a substantial part of anti-corruption regulation for government officials around the world. In the case of public servants, it is very clear what public interest refers to and what the personal interest is, so that any outside person can describe the conflict. However, in the case of companies, the situation becomes opaque, because the corporate interest cannot be interpreted as public (in the sense of social welfare) and because a majority of firms are born from the founder's entrepreneurial vocation. At least in the first years of a company's existence, we cannot make the difference between the firm's interest (its objective of making profit) and the investors' interest (of obtaining dividends and other pecuniary benefits). In the case of family-owned enterprises, when one member of the family is also the CEO, pecuniary and managerial interests are intertwined, so that CIs are improbable and largely ignored.
When do CIs in business appear? We can assume that they appear gradually, along with the growth of the firm. It is well known that the size of the firm is linked to a specific degree of separation of ownership and control, which is a special case of agency conflict (Jensen & Meckling, 1976). Owners of a firms have a specific pecuniary interest (to protect their investment) and professional managers have a large array of interests, of which some are financial (in terms of compensation), and others are linked to power and prestige. To align the often-divergent interests of the agent and the principal is the main objective of corporate governance theory and practice. There are two main mechanisms of achieving this task: compensation and external supervision. Both of these tools are external ways of dealing with potential CIs, but the philosophy of ethics recommends a third solution: selfsupervision.
The seminal work of Luebke (1987) describes the parameters of a CI, which are intimately linked to agency theory: the existence of the bearer of the CI (the person empowered to act in a fiduciary role), the marks of the conflict (when the fiduciary party has an interest that may be adverse or is likely to become adverse to the interest of the principal or the employer), and the moral prescriptions for dealing with a CI. Here we arrive at the concept of self-supervision or self-regulation. Luebke considers that there is nothing morally wrong with being in a moral dilemma, but it is wrong to willfully enter a conflict of interest or to refuse to resolve it. Also, it is morally wrong to consciously attract someone into a CI.
The recent literature on CIs in business is remarkably thin. Even though one could find contributions focusing on CIs in sport management (Sherry et al., 2007), auditing services (Moore et al., 2006), director stock compensation (Dalton & Daily, 2001), board responsibilities (Sherry & Shilbury, 2009), and institutional investors (Ivanova, 2017), research has ignored the magnitude of this ethical problem in the business world. The original contribution of this paper lays in the classification of CIs in business and on a model predicting when and if a person may enter a conflict of interests. The perspective drawn from the cognitive psychology literature supports the current findings and extends the discussion of how can people avoid or resolve CIs.

Conflicts of interest and the corporate objective
Before attempting to classify the components of CIs in a business context, one more step is necessary: to define the conceptual dimensions of CIs. I will discuss several terms: corporate objective in view of shareholders' interests, the fiduciary duty of corporate employees, and the impact and severity of CIs.
In very broad terms, the corporate objective is value maximization, which works in parallel with social welfare and the accountability of managers and directors (Jensen, 2001). Value maximization per se is morally neutral, it is not a wrong or a right to have profit or to record losses. However, the correlates of economic activity are not morally neutral, and this is the main claim of stakeholder theory, which argues that managers should make decisions so as to satisfy or at least not to affect the interests of all stakeholders in a firm. Business ethics is an avenue of action which ensures that all parties' interests are respected, if not maximized.
When talking about the members of an organization, the analysis of CIs is focused on the reciprocal argument: the fiduciary duty of employees is to take into account the interests of the corporation (and eventually those of the investors). At a very basic level, even a false sick leave may have serious consequences if the person taking days off has responsibilities for signing new contracts with important customers. Thus, employees are invested with trust that they will not harm the company by affecting the corporate objective. Officers are also hired by the company: they are nominated by the Board of Directors and their appointment voted by the Annual General Meeting. The social responsibility of an organization is to protect the interests of its members, and the fiduciary responsibility of employees, officers and directors is to work productively and act with integrity in the interest of the firm.
Violations of corporate policies may lead to CIs of varying severity. The rule-ofthumb will be that the higher the responsibilities of the person involved in a CI, the higher the impact of the ethical violation. However, we can assume that certain CIs can have more severe consequences than others, all things being equal. Favoritism can have disastrous consequences when it happens at executive level, but breaches of confidentiality over intellectual propriety are major risks irrespective of who is the person which leaks inside information. It is safe to say that strictly internal CIs have lesser impact than cases which involve competitors, contractors or the financial market, the latter posing significant threats to corporate reputation.
However, CIs are rarely strictly internal, so it is impossible to create a hierarchy of CIs based on their assumed severity.

Methods and results
I have conducted an interpretative analysis of codes of conduct belonging to the world's top 10 largest companies by market capitalization at the end of 2016 (World Economic Forum, 2017). The codes analyzed for this purpose belong to: Apple -AP 1 , Alphabet (Google) -GO 2 , Microsoft -MS 3 , Berkshire Hathaway -BK 4 , Exxon Mobil -XO 5 , Amazon -AM 6 , Facebook -FB 7 , Johnson and Johnson -JJ 8 , General Electric -GE 9 , JPMorgan Chase -JP 10 (access date for all documents: March 3, 2017).
The main topics discussed under the heading "Conflicts of interest" in the analyzed codes are summarized in Table 1. The final row of the table contains the word count for the respective section of the code of ethics. The information is constantly overlapping and the definitions are usually quite similar. However, each company chooses to classify the relevant CIs in differing subchapters, mainly connected to the specifics of the industry and the scope of the code. The categories in Table 1 are a condensed version of the classifications offered in the codes of ethics. The discussion does not focus on CIs specific to the Board of Directors such as: related party transactions, pending or threatened litigation between the directors or officers and the corporation, the actual or potential use of confidential information by the director or officer for unauthorized purposes, insider trading, and outside directorship with unaffiliated parties (either for-profit or nonprofit). The ethical principles applicable to ordinary employees are also relevant for directors and officers, with the observation that the latter have access to much more resources and information than other members of the organization, which expands their responsibility. However, at Board level there are better safeguarding mechanisms, such as the Audit Committee, the existence of independent directors, the involvement of major blockholders and internal whistleblowing.
A visual conceptualization of CIs in business is offered in Figure 1, and it is an original aspect of the present research. The instances of CIs are written in italics, and the employees are at the center of the diagram, because they are the main addressees of code prescriptions. Productivity, asset protection, confidentiality and impartiality in business relationships are the four corporate values which are affected by CIs. On the vertical axis of the diagram one can find the groups of persons usually connected with CIs. The model is descriptive and synthetic, and other combinations or types of CIs are also possible. However, most codes of ethics cited in the following section are centered on these particular aspects of the concept, and mostly on receiving gifts, bribes and commissions. The model omits the CIs induced by our employees in other individuals, such as when offering bribes to public officials.
The following subsections provide a synthetic discussion of the most relevant CIs pertaining to the following areas: employment, contracting, corporate assets, insider trading and personal investments, competition and corporate image. There is a general consensus between codes regarding these ethical aspects, both in the spirit and the letter of the provisions. It is very probable that all analyzed companies also have internal (nonpublic) rules for treating CIs on all the above topics. The following discussion focuses on publicly available guidelines and aims to identify the conflicting interests while evaluating the severity of each case. However, the ethical arguments need to remain in balance, because not all CIs are critical to the profitability of the company, and not all circumstances turn into CIs. The utilitarian principle (Sinnott-Armstrong, 2015) that "all is well when ends well" should be a moderator of the estimated weight of the consequences.

Employment
Generally, the CIs linked to employment are internal to the company, but they may have major repercussions, albeit indirectly, on the efficiency and profitability of a company. There are four main aspects of CIs in employment, i.e. nepotism or favoritism in the hiring process, outside employment, biased supervision of family members and romantic relationships in the work place.
Nepotism is favoritism shown to relatives, by giving them positions in spite of their competencies. Family members include persons such as a spouse, domestic partner, parent, sibling, child, grandparent, grandchild, aunt or uncle, niece or nephew, cousin, stepchild, stepparent, or in-law. Incompetence destroys value, and nepotism is usually associated with incompetence. Being a member of an organization creates the normal drive to support one's family using the resources of the organization. It is not uncommon for people to try to introduce family members into their organization. This way, undue pressures are put onto the human resources department or on managers to create new jobs, to alter the hiring process or to favor one candidate over another. If we apply the deontological view, nepotism is bad irrespective of the characteristics of the person receiving the favors. In the teleological (utilitarian) view, nepotism is bad only if the company hires a "bad apple". Codes of ethics adopt the deontological view and prohibit any involvement of company associates in the hiring of their family members, although HR departments may accept recommendations for open positions.
Outside employment is mostly prohibited by corporations. They maintain that any other business activities outside working hours will disrupt the quality of the work performed by the employee for the corporation. Some companies do permit other employment or activities, provided these do not conflict in any way with corporate interests. Of course, this is acceptable as long as the outside employment does not involve a supplier, a competitor or customer. In this case, the employment situation creates at least the appearance of a CI because of the breaches of confidentiality that might occur and because of the threat of favoritism in contracting.
Serving on the board of directors for another firm may be an instance of a CI, especially if that firm supplies goods or services to the company or purchases goods and services from the company. Clearly, serving on the board of directors of a competitor is a clear violation of business ethics. While serving on a board of directors for a non-profit organization is generally encouraged and does not require prior approval from the Ethics Office or Corporate Governance Committee, it is important that this position does not interfere with the employee's ability to perform job duties.
The supervision of family members in the workplace is a classic example of a CI, which instantly affects all the relationships in the workplace and throws a shade of doubt at all the decisions taken by the supervisor. Basically, no team member in a department trusts that the personnel decisions taken by the team leader are not biased, if the latter is also supervising a family member. Most of the time, these biased decisions imply the unfair distribution of rewards, the overly positive evaluation of employee performance or the allocation of better-paid work (Prendergast & Topel, 1996). Moreover, nepotism create job stress in the workplace and this increases dissatisfaction of the staff about their organization (Arasli & Tumer, 2008).
The above situation is perfectly replicable when the manager and a team member are involved in a romantic relationship. The disruptions of workplace morale and the decrease in efficiency may have a major negative impact on productivity and the quality of collaborations. Generally, supervision of family members (here including romantic interests) may be a consequence of favoritism in the hiring process. The consequence of this type of CI is that, once a case like this is tolerated in the organization, it gives proper ground for other employees to follow. Nepotism is a type of "ethical gangrene", because it may slowly and insidiously alter the ethical tissue of an organization, and may become almost impossible to address and eradicate after a certain point.
Romantic relationships, besides the obvious CIs arising from supervision, can be a fertile ground for harassment and undue pressure. Relationships may grow harmoniously between two people, but in cases of breakup, professional collaboration may be deeply affected and productivity will sour. The company may address this situation before it leads to employment termination, by offering alternative job positions, in separate departments, with as little as possible interaction that may affect the professional involvement.
Companies can be very specific about one "asset" in particular: the working time of its employees. Many companies do not permit internet access to social media websites, gaming activity, chatting outside work interests and other typical procrastination attitudes. The rationale behind these measures is that time wasting affects productivity. This is an argument that goes both ways: taking it face value it appears true, because the employee does not perform any relevant task while playing or checking her Facebook; however, this argument does not account for two things: the lack of tasks to be performed and, at the opposite side of the spectrum, the fatigue arising from continuous concentration on work tasks. The full-time employment arrangement, while providing good control over the working time, does not take into consideration the gaps in productivity (idleness), the normal working rhythm of different persons and the time perspective of the individual inclined to procrastinate (Gupta et al., 2012).

Contracting
To initiate a commercial relationship with family members in the name of the company is a clear instance of self-dealing. It has the appearance and the substance of a CI, because the company's business is detoured towards increasing personal wealth. It is different from facilitating employment for family members, because in commercial deals the resources involved are greater and the risk of making a "bad deal" are also much greater.
Hidden commissions are a form of bribery, and are also referred as kickback payments. They occur when a previously contracted agent pays part of an excessive service fee back to the employee who has signed the initial deal. Usually, these repayments are kept secret, because this is a private deal between the agent and the purchasing manager. The excess fee is taken from the company's funds, so the employee is actually tunneling corporate funds into his own pocket. These arrangements are not only unethical, but they can be brought to court if the company cannot recover its damages.
Business gifts are extensively discussed in all codes of ethics scrutinized in this paper. Meetings for signing business deals may very well be accompanied by gifts, entertainment and other courtesies. There are etiquettes and local customs when it comes to business gift giving, as in Japan, where gifts generate an obligation. Arunthanes et al. (1994) discuss gift giving in high context cultures with implicit, non-verbal, contextual communication styles (e.g. Japan, Arab and Mediterranean countries) versus the low context cultures which rely on promptness, explicit contracts and formal negotiations with a legalistic orientation (e.g. the U.S., Germany and Switzerland). The codes of ethics belonging to European and U.S. companies have very strict policies related to gifts. In some cases, gifts are prohibited altogether, while in others, they are acceptable if the gift is not cash or a cash equivalent, has a legitimate business purpose, is of nominal value (generally under $100 USD dollars) and is infrequent, and if the employee's division or function does not have a "no gifts" policy in effect.
Giving or accepting valuable gifts or entertainment might be construed as an improper attempt to influence the business relationship. This is especially true when it comes to bidding or contracting with suppliers. However, accepting business meals from suppliers is common, but should be modest and infrequent, and never during contract negotiations. Impartiality is of paramount importance for the contracting procedure, and all codes agree that receiving gifts, meals and entertainment (e.g. party invites, sports tickets), or travel and accommodation expenses affect at least the appearance of impartiality.
One more relevant aspect related to contracting involves authorization or explicit approval, which must be obtained before carrying out investigations, negotiations, tendering/bidding, providing guarantees, issuing attestations, or signing contracts. Entering into unauthorized deals in the name of the firm can have potential disastrous consequences, from financial losses to tarnished reputation. This situation is different from the hidden commission scenario, because in the latter example, the manager is authorized to close the contract, but he later alters the financial terms of the deal.

Corporate assets, products and services
Corporate assets include: physical assets, such as office furnishings, equipment and supplies; financial assets, such as cash, securities and credit cards; technology assets, such as computer hardware, software and information systems; information assets, such as intellectual property, including information about products, services, systems and other data; and the corporate name, its brand and our customer relationships. Companies generally need to be very protective of their assets, products and services, and especially of their intellectual capital. One rule in this respect is that innovation produced by employees as part of their employment becomes the property of the company. The situation may become critical after a person has left employment: the company must ensure that proprietary information is not disseminated without authorization, especially in the advantage of the future employer (which may also be a competitor). For this reason, managers know that people carry sensitive information and must ensure employee loyalty.
From the very mundane act of printing personal documents on the company printer to using the company name for self-dealings (such as paid public speaking), any such instances may qualify as CIs. A firm needs to strike a balance between what it offers to employees and what is restricted to use. By allowing some assets, products or services to be used for free or in better terms, a firm can lower the propensity of its employees to deplete company resources.
Companies identify a CI whenever an employee uses the company's products, services or information in a way that improperly benefits a friend or family member, or creates the appearance that any employee has an unfair advantage over outside users. For example, an employee should never approve corporate accounts, services or credits for himself, her friends, or family members. The approval of accounts, services or credits is very traceable because integrated information systems keep logs and records of every decision, and it is easy to implement formal safekeeping in this respect. However, the situation in which a manager may grant his employees unfair advantages, in return for other types of favors, is a clear statement of a conflict of interests which is hard to investigate and which may harm the organization on the long term. For many people, helping their peers and family members is more important than loyalty to the employer, which may give rise to a conflict of interest.

Insider trading and personal investments
Insider trading is a strongly regulated criminal offence in all countries with a functional financial market. However, most codes of ethics also address it because employees are usually tempted to do just that: to use material, non-public information obtained in the course of their work to trade the company's securities or instruments of another entity with which the company the business, such as a supplier or business partner.
"Tipping" is also a violation of the code and state law. Tipping arises when material, nonpublic information about a company is disclosed to someone else, and that person trades stock of that company while in possession of the respective information. Moreover, the prohibition on tipping also covers discussing sensible information with other company employees, unless they have a business need to know. Tipping is an offence even if the employee did not personally make any trade based on the respective piece of information.
Information is "material" if it can convince a reasonable investor to buy or sell; information is "nonpublic" if it has not yet been released through the official communication channels and if the market has not yet absorbed the information in the valuation of publicly traded securities. In other words, insider trading affects market efficiency and leads to illegal financial gains for the insider. Examples of material, nonpublic information include: nonpublic financial results, development of a significant new product, unannounced mergers or acquisitions, pending or threatened litigation, or advance notice of changes in senior management.
Employees of financial institutions have a large array of investment opportunities on their hands and the relevant knowledge to buy or sell securities for their own portfolio. In this respect, CIs may arise whenever an employee: resorts to shortterm or speculative trading; makes trades in possession of confidential information relating to those securities; goes beyond her financial means (and borrows funds from friends, co-workers or unauthorized money lenders); or invests in competitors, such as banks, lenders, private equity firms, asset managers, depository institutions, credit unions, investment banks, insurance agencies, and securities brokers, dealers and underwriters. These restrictions usually apply to clients or suppliers: while negotiating a contract with a third-party, employees are prohibited to invest in the securities of that entity. The rationale behind these rules is that any improper handling of personal investments could undermine the credibility of the employee (as a financial expert) and the company's.

Competitors
Leaking information to competitors either through outside employment or industrial espionage is one of the most serious blows to business profitability and is generally followed by employment termination and legal actions. Most employees sign confidentiality agreements upon hiring and have explicit prohibition of contact with competitors. Examples of confidential information include plans, earnings, business forecasts, financial forecasts, market share, costs and margins, distribution methods, competitive bids, discoveries, technologies, and personnel.
Companies are especially wary about agreements between corporate employees and a competitor about aspects such as pricing, bidding, deal terms, wages or the allocation of customers and markets. Also, contacts with competitors readily create at least the appearance of a conflict of interest. Another side of the coin is when the company obtains unauthorized information about a competitor, through illegal channels (leaks). Although highly advantageous to the company, this is still a very serious CI. If fair competition is at the heart of the business, the CEO should inform the Board of Directors, and together they should discuss how to address the situation and not use the respective information.
However, there are cases in which employees themselves compete with the company, as in the case in which an employee is developing his own products or inventions that relate to the company's existing or anticipated products and services, or are developed using corporate resources. These situations are likely in high-tech companies where programmers may choose to develop for their own personal projects, which may be based on proprietary knowledge or may use company development tools. Another type of CI appears when the employee wants to sell his own product through the official corporate online store. These cases can be easily resolved if the company chooses to incorporate that side-project into their strategy or future products.

Corporate image and reputation
Corporate image is an asset, which includes the name, logo, brand, as well as a reputational componentthe public perception of that company (Gray & Balmer, 1998). In the social media (here including online technical forums), employees of high-tech companies can be very active discussing the company's products. Overly positive product reviews promoted by company employees may actually create a negative stream of comments that affect company reputation. The reverse is also true: being excessively critical of a company product may create a conflict between the employee's duty of loyalty and his interest of self-promotion. Fairness in judgement is the best choice, while any employee must be aware that he is always publicly perceived as a representative of the company, and anything he says can be viewed as an official statement.
Corporate image is involved whenever an employee (including officers and directors) are engaged in public speaking that relate to the business or their own interests. Employees must obtain authorization from their managers because the public tends to associate the speaker (and the content and quality of her presentation) with the corporate image. The same restrictions apply to publishing articles or books where the author identifies himself as an employee or officer of the company. Such publications can be technical (about the company's products or services) or general business and management literature. The real danger concerning such publications is that the employee could break confidentiality agreements related to disclosing business processes and strategy.

An overview of content analysis
As a conclusion to this section, I have summarized the results of content analysis in Table 2. Having put corporate interest and self-interest in parallel gives us a very relevant outlook on the scale and diversity of ethics violations falling under the generic term of conflict of interest.

A model for dealing with conflicts of interest
Codes of conduct usually describe different types of conflicts of interest, but they do not explain why they are so. The present paper has aimed to fill the gap and create this double perspective: the corporate interest and the self-interest of the employees. As stated in the Introduction, these conceptual separations are often blurry. In many cases, especially for small firms, there is an overlap between the company's interest and the personal interest of the employees. The alignment of these two types of interest is beneficial to the firm because it keeps the employee loyal and devoted to the economic interest of the firm. This situation occurs whenever the economic resources of the firm are confounded with the personal resources of the employee.
When the company's resources (assets, employment, working time, innovation, business relations) become impersonal, a classic governance problem appears: the separation of ownership and control (Berle & Means, 1932). All employees have some limited degree of control over company resources, with the CEO having the most control. At this point, the corporate interests and the personal interests become divergent, and conflicts of interest may appear. From this point, the CI is a latent state: circumstances can trigger CIs, but most of these are under the control of the employee.
People may enter into CIs for three reasons which act concurrently: (a) out of ignorance: they are not aware of the rules, although most people would have the intuition of a conflictual situation; (b) because self-interest is too strong and the situation is advantageous to the employee; and (c) because it is common practice (the social norm). In brief, people usually estimate costs and benefits of entering a CI, even if they may not perceive it as a CI or may not be totally aware of the costs (administrative or legal consequences). In general, this estimation is biased towards the benefits of a particular situation, assuming that self-interest is an automatic response. This argument is graphically represented in Figure 2.
Considering that the CI is a romantic relationship between the supervisor and a team member, the cost of resolving a CI would be employment termination for both of them. Even if "love is blind", a romantic affair in the office can also be avoided after a mature conversation, if it would be seen (by the code of ethics or other people) as a conflict of interest. In case (a), the team member has monetary incentives to not avoid the CI because he would get better evaluations and subsequent bonuses from a biased supervisor. For the manager, the monetary incentives are replaced by sexual or emotional gratification (which do not have a value function attached). In case (b), there are no monetary incentives to enter the romantic relationship, because the fear of employment termination is greater than the potential rewards of this relationship. In cases (c) and (d), the benefits are varying over time. If the beneficial outcome for the employee is increasing proportionally to the amount of time passed (c), there is a certain point at which the benefits extracted from the CI surpass the costs of resolving the CI. If the benefits are decreasing, the employee will be tempted to resolve the CI and to end the relationship or to move to another department (d). One may argue that this view of starting or ending romantic relationships (the object of the conflict of interest) is overly cynical, but the fact that people take advantage of such relationships to advance their careers proves that this line of reasoning is embedded in the motivation of engaging in intimate affairs at the workplace.
Cognitive psychology research has tested the hypothesis that humans are rational beings guided by careful reasoning over costs and benefits of future actions. Schneider and Shiffrin (1977) have introduced the notions of automatic and controlled processing, which can be applied to ethical reasoning as well (Moore & Loewenstein, 2004). Automatic decision-making assumes that ethical decision are taken based on preformed patterns of action, which rely on a attitudinal basis regarding one's social environment. Controlled decision-making appears on demand, it is resource-intensive and seeks to create new patterns of action or to change existing ones.

Figure 2. Benefits and costs associated with conflicts of interest
The moral imperatives that we internalize through childhood can be seen as instances of automatic ethical reasoning. For example, we normally do not question the imperative "do not steal" every time we go into a store to get food. In contrast, controlled decision-making is set to be triggered whenever we are in an ethical dilemma. At this point, we balance diverse possibilities of action, either through the deontological ("it is forbidden to do that"), the utilitarian ("it is ok to do that if everyone will be ok") or the non-ethical view ("it is ok as long as I do not get caught"). Solving ethical dilemmas generally require a substantive effort, a time perspective and the anticipation of consequences, in a probabilistic manner. This process is costly and thus infrequent. A much more easy way is to adopt already available ethical solutions and to apply them to cases following a certain blueprint. The purpose of codes of conduct is to offer such ethical solutions for corporate employees to adopt as default patterns of action.
Traditional economic models of rationality would assume that people can perform optimally in playing dual roles, making unbiased judgments when it is in their interest to do so. Moore et al. (2010) have shown that the psychology of conflict of interest is at odds with the way economists routinely think about the partisan thinking and acting. Once primed into a partisan role (assuming a certain ethical stance), people find it difficult to extricate themselves from this role and make a neutral judgement, even if they are expressly told to do so. Moreover, they honestly believe that they are acting impartially, although the empirical evidence shows otherwise. These results show that automatic decision-making takes precedence over controlled reasoning, and that preformatted solutions work more efficiently (effortlessly) and unconsciously. The cited paper shows that unethical (biased, partisan) decision-making, once primed, becomes the automatic way of action and is consolidated. Well-thought codes of conduct and ethical training are an effective way to automatize the decision-making process in the right direction, and to avoid conflicts of interest as much as possible.

Conclusions
From a psychological perspective, self-interest is an automatic response, in line with an evolutionary self-preservation instinct. Automatic responses can also be formed, through operant conditioning (reinforcement or punishment) or through priming (exposure to one stimulus can influence the response to another stimulus). From a utilitarian point of view, entering or avoiding CIs is planned behavior and it relies on an evaluation of costs and benefits in a time perspective. Refraining from using corporate resources for personal gain is a calculated, difficult and therefore ethical response, which demands education and the existence of explicit guidelines. These two perspectives are not contradictory: they merely describe the "dual processing" mechanism of the human mind in relation to an actual or potential ethical dilemma. Or should we say the "interest serving" dilemma?
Self-interest is identical to the economic interest of the firm only in the case of the entrepreneurial venture. In any other situations, personal interests diverge (sometimes significantly) from the corporate interest. Companies have several possibilities to prevent conflicts of interest: protecting data privacy, implementing stricter procedures for employment and contracting, setting up detailed supervision criteria and peer review, establishing quantitative measures for gifts and entertainment (either given or received), creating an Ethics Committee and providing ethical training to all employees. Most of all, companies foster an ethical climate if the managers act ethically at all times and offer a personal example of irreproachable conduct.
Preventive measures are much easier and less costly to implement than corrective actions with regard to conflicts of interest. However, critics may argue that opportunistic behavior is likely to make agents run afoul of principals, which warrants governance mechanisms to align the interests of agent and principal. Thus, self-supervision may become a breeding ground for taking advantage of principals. To be consistent with agency theory, self-supervision could also be compelled by an exogenous variable that contributes to discipline the agent (a case in point could be the reputational implication of failing to self-supervise).