Corporate Codes of Ethics-The How Factor

Although recognized as a key factor for its effectiveness, the adopted process for the development of a global corporate code of ethics, is one of the least documented aspects by Academia. The code is often created at central level and then sent for adoption and implementation to the different branches within the organisation, and the fact that they are not involved in the development process, frequently elicits resistance and compromises effectiveness. This was initially the case with the company of this study, where it was found that business units from different geographical locations had gradually made adaptations to the original code of the Group, which no longer remained the same across countries, neither in text nor in form, which led to the code review. Developed through an action-research methodology, this case study describes the review process which included 30 people from all the Group companies. All defined goals have been achieved. The different companies’ representatives consider that the participatory methodology was a key-factor to the general acceptance of the produced document. Three years after the code release, its efficacy was also evaluated with the Ethics Ombudsman who presented possible success indicators.


Global Corporate Codes of Ethics and Relevance of The Development Process
A corporate code of ethics is a formal document developed by a company to express its commitments and the commitments of its employees "at least its managers and employees towards one another, the company, external stakeholders and / or society in general" (Kaptein & Schwartz, 2008, p. 113). It clarifies the objectives of the company, its standards and aspects related to ethical behaviour; it regards the company's ethical responsibility in any part of the world it operates, even if it operates in a large number of countries. Take the case of Volvo -documented by Helin and Babri (2015) -which has production in 19 countries, employs more than 100 thousand people, operates in more than 190 markets and has a unique code of ethics, valid anywhere.
In fact, most companies operating in different markets develop a global code or "At least some standards of ethics that are binding on all employees worldwide" (Talaulicar, 2009, p. 349) because it is viewed as "A much recommended management instrument" (Kaptein, 2004, p. 13) to ensure or show that an attempt is being made to guarantee that the company assumes its responsibilities wherever it operates. Moreover, since companies use the same management model for their operations wherever they may be present, having the same universal approach to ethics would seem to make sense. "A global approach seems to fit the strategy of global firms because they expand internationally using a specific business model as a proven standard format" (Talaulicar, 2009, p. 350).
We use the expression code of ethics here, although many companies use the word ethics and conduct for the same type of document. In any case, there is formally a difference between the two concepts that may have an influence on its acceptance by employees in different countries. The codes of ethics tend to be more inspiring, axiological, more focused on principles, whereas codes of conduct are essentially a set of rules to be followed, in other words, they refer to obligations "whereas principles obligate to approximate ideals. More specifically, rules tend to be more concrete because they explicitly define which conduct the norm addressees have to apply. In contrast, principles remain more general" (Talaulicar, 2009, p. 356).
These global codes of ethics generally address a range of issues that do not seem to vary greatly depending on the company or its location (Kaptein, 2004), for example: relationship with stakeholders, transparency, honesty, justice, discrimination, or fraud; appear in almost all codes but, in reality, they subtly vary according to the nationality of the head office (Scholtens & Dam, 2007, p. 274). This variation may be the source of some of the implementation problems, how the code is accepted and how it is adopted by employees from different geographical locations.
Harvard Deusto Business Research. Volume X. Issue 1. Pages 156-175. ISSN: 2254-6235 Let us begin with the content, what is explicitly written: Innovation and creativity, for example, are respectively mentioned 73% and 46% more in Asian codes than in American and European codes. By contrast, the value responsibility / conscientiousness can be found just as often in American (30%) as in Asian (35%) and European codes (33%). (Kaptein, 2004, p. 27) These are perhaps subtle differences, but they have been pointed out in a study that analysed the codes of the 200 largest companies worldwide. Naturally, culture differs from country to country, and certainly from continent to continent, and the same applies to codes. This corresponds to a need to show a global "voice" for the companies, rather than a true reflection of a unique way of feeling and being.
As for the way in which it is expressed, there are differences in many cases that may have an impact on the code acceptance. A good example is the case described by Helin and Sandstrom (2008) regarding the code of ethics implementation of a US company in a branch in Sweden. According to the mentioned authors, the code arrived in Sweden accompanied by an order to: "implement it and make everybody sign it" (Helin & Sandström, 2008, p. 283). This message, which certainly does not differ from the messages sent by many other parent companies to their different subsidiaries, may in itself give rise to a non-acceptance and even a total rejection of the code: "The receivers resisted the code by amplifying the importance of national identity. Rather than stimulating a discussion on ethics that might have strengthened the ties between the parent and the subsidiary, the outcome of the code implementation had the opposite effect" (Helin & Sandström, 2008, p. 281).
In the previous example, the front reaction of the Swedish subsidiary's CEO helps us clarify the issue. He refers to the code as follows: "The rules presuppose that people are stupid. They treat people as less intelligent. Then people become pissed-off and then it can have a reverse effect" (Helin & Sandström, 2008, p. 285). And he is not the only one expressing disagreement, the production director accuses the strangeness of it, and in other words, says that the content does not truly reflect him as "It felt as if I was not at home when I read the rules. It was all about the US" (Helin & Sandström, 2008, p. 285).
These reactions corroborate the thesis that business ethics must be approached in context, because it does not arise in an isolated and unconstrained way: "It takes place in a social and cultural environment that is being governed by a complex set of laws, rules and regulations, formal values and norms, codes of conduct, policies" (Scholtens & Dam, 2007, p. 273). This influences the way content, form, and even tone can be interpreted: "For example, in an Asian country, a typically American code of conduct could be regarded as too comprehensive and forceful while the business principles of an European company could be viewed as too ambitious and abstract" (Kaptein, 2004, p. 27). That is why many schools of thought emphasise "the necessity to take culture into account when dealing with ethics in business" (Helin & Sandström, 2008, p. 281).
That is, as Kaptein and Schwartz (2008) point out, the word code has at least two meanings: the set of rules to which we normally associate the word, but before that, the code refers to something that can be deciphered in different ways in different cultures. The code can be seen as a "system that gives meaning to a series of symbols, signs, or signals such as Morse code, the binary code, and bar codes" (Kaptein & Schwartz, 2008, p. 112).
It is a paradox. On the one hand, uniformity is not necessarily an added value. As (Kaptein, 2004, p. 24) says, the diversity in the content of corporate codes (also within countries) is not necessarily a negative sign. To have a different code of ethics from country to country, or from culture to culture, could even make sense: "When firms go global, they face ethical complexities as they operate in different legal and cultural environments that may impact the admissibility and appropriateness of their approach to institutionalize and implement corporate ethics" (Talaulicar, 2009, p. 349). On the other hand, as Moreira (2002) points out, local culture does not have to determine a code, because that culture may contain abuses and a code must be deduced from "normative principles that do not say how the majority acts -that would be sociology -but how we should act" (Moreira, 2002, p. 73).
But the truth is that, for the sake of security, or apparent ease of management or even obligation, most transnational companies choose to have a single global code of ethics in all their places of operation, and those that did not adopt this solution yet, are getting there (Kaptein & Schwartz, 2008, p. 111), by pressure of their stakeholders or forced by law. The challenge is to make a single code, but one with which the stakeholders of the different cultures where the company operates feel comfortable.
Taking culture into account can begin at a very early stage of the code creation process, including reflection and dialogue with stakeholders. This can produce significant benefits, especially in the acceptance of the created code, which will be less likely to be rejected in the subsidiaries. This is especially important if the goal of the code development is to promote change in the organisation. The necessary dialogue for participation may already be part of the intended change and alignment. "Ethical leadership and open discussions of ethics in the organization contributed to increased ethical behavior" (Stevens, 2008, p. 604).
The relationship between the code of ethics development process and its impact is recognised. Kaptein and Schwartz (2008) pointed out the importance of creating code supporters. This creates awareness and appropriation by employees, which in turn makes the code real.

Relevance of This Study
Most studies on corporate codes of ethics have focused on content and effectiveness, while disregarding the employed process for its creation (Kaptein, 2008); "(e.g., Cleek & Leonard, 1998;Davis, 2000;Frankel, 1989;Kaptein, 2004;Verbos et al., 2007)" in (Messikomer & Cirka, 2010, p. 55). The aforementioned process is not something that should be disregarded, or which has no influence on content. In fact, Erwin (2011) stated that the adoption of the development process has a direct implication on the adequacy of the code content (Erwin, 2011, p. 537 Identical vision is advocated by Kaptein and Schwartz (2008), who stated that similar codes may have different results, depending on how they were developed and, as a result, "the effectiveness of a code can diverge even if two companies have an identical code implemented in an identical way" (Kaptein & Schwartz, 2008, p. 119). Messikomer and Cirka (2010) even argue that it is difficult to establish a boundary between what constitutes the code-making process and the product itself, i.e., the code resulting from it.
The process has a direct influence, for example, on the identification of relevant issues, which would be difficult to achieve without a participatory method (Messikomer & Cirka, 2010, p. 59). In this regard, the same authors who developed a case study on the process of elaborating a code in a professional organisation, concluded that the process is the area which has been less studied. They say that virtually no empirical work on this topic exists, aside from a few limited cases (Farrell et al., 2002a;Frankel, 1989;Gotterbarn, 1999;Kaptein & Schwartz, 2008;Kaptein & Wempe, 1998;Nijhof et al., 2003;Sinclair, 1998) in (Messikomer & Cirka, 2010, p. 58).
Thus, this article aims to create awareness in the aforementioned understudied critical area, through the dissemination of the code of ethics review process of a transnational company, where we played the role of actor and researcher. That is, one of the authors was part of the committee, and also a researcher on the subject.
The process documentation and narrative of this case study aims to show and validate the relevance of the participatory method in constructing the code, its acceptance and, ultimately, its adoption, i.e., being incorporated by the organisation.
It refers to a qualitative, exploratory and descriptive case, in which the purpose is not only to document the process, the methods used, the doubts and emotions felt by the team during the process, but also to propose generalisations based on literature lessons and on the obtained experience and reflection.
The relevance of our case, a single case, is, as Yin (2001) pointed out, the fact that it can be "used to determine if the propositions of a theory are correct" (Yin, 2001, p. 62). Therefore, the thesis is that the participation on the development process not only facilitates the processes of codes of ethics appropriation in different branches of transnational organisations but is also crucial for the quality of the document produced, better, the quality of the result. Additionally, this particular approach has been identified "as advantageous in addressing the complexity of path-dependent explanations" (Roque et al, 2020, p. 277).
Currently, more and more companies are acting on a global scale and are concerned about the standardisation of values and practices. Therefore, testing methodologies that allow for a change from imposed codes, possibly only elements of a formal ethics, to codes adopted by employees, with the potential of actually becoming part of the day-to-day within businesses, takes on special importance. To have people convinced about the code is fundamental for its existence, because the diffusion of the ideas contained in it is not done per se, "it is people who pass them on to each other, each one translating them according to their own frames of reference" (Czarniawska, 1997 p. 95). In other words, individuals ultimately are the ones who truly make the code alive and a relevant element for the organisation, a true contribution to a change in behaviour and to the ethical climate.

Methodology
The structure of our case involved a storytelling approach, centred on the ethics ombudsman and the code-development team. This approach aims to generate a narrative that helps to understand the process, challenges, solutions, obtained results, and the way that the process was felt, lived and reflected by those who participated in it: "The narrative is a way for us to make sense of our experiences to ourselves and relate those experiences to others" (Poulton, 2005, p. 1). In summary, we were interested in studying the impact of the process on the product.
Our narrative follows a chronological sequence: beginning with the background and environment, it presents the process and the evaluation developed by the participants of the various geographies upon launching the code; and concludes with the vision of the process, and the evaluation of the code effectiveness, by the company's ethics ombudsman, three years after its initial disclosure.
This code was the company's second code of ethics, and we used excerpts from an interview conducted to the only element present in the team that developed both versions of the codethe original 2005 code and the revised one. Different business units (BU) team members' testimonials of the review, extracted from the ethics ombudsman's report, were also used, as well as an interview with the company's ethics ombudsman (EO) which, although carried out retrospectively (three years after the code review process), will have excerpts included in the narrative of the case itself.
The interview with the ethics ombudsman represents the company's view after the test of time and is very relevant because it allows us to identify gaps and success indicators, for which there was no clear perception at the time.
Some excerpts from the interview are also included in the discussion, since part of this final interview also dealt with concerns and relevant aspects for which we still do not know the answer to and believe they are important to delve deeper.
The case study was created based on an action research strategy with one author participating (as an academic and as a consultant) in the process. The first objective of this participation was to support the development of the code itself: "to make action more effective while simultaneously building up a body of scientific knowledge" (Coghlan & Brannick, 2005, p. 4).
The fact of having integrated the entire process allows us to develop this narrative with a perspective. The researcher supported the search with external information to the company, The objectiveness about the results obtained and about the way in which the process was carried out is ensured by the statements from other participants and, above all, by the final texts of the codes of each company of the Group. However, in the course of this narrative, references to the researcher´s feelings often arise, which are, some of them, corroborated by the words of the ethics ombudsman, in the excerpts of the interview.

Company Presentation
The company is a global economic Group present in 14 countries, employing roughly around 12 thousand people. It is recognised worldwide in the field of sustainability as one of the reference companies, with tenth consecutive presences in the index the Dow Jones Sustainability World Index (DJSI), some of them as sector leader.
In addition to being evaluated annually by the SAM methodology of the DJSI indexes, it is worth highlighting, within the field of ethics, its presence in the FTSE4Good index since 2011, and in the World's Most Ethical Companies ranking of the Ethisphere Institute since 2012.
In 2005, the Company launched its first code of ethics (Code). In 2009, after the Regulation of the Code of Ethics' publication (Regulation), an ethics ombudsman was appointed, and employees underwent a broad training (Figure 1). The company was aware that "when the implementation of a code is not supported by other instruments, it has negative effect on employee's perceptions of ethical behavior in the workplace" (Kaptein & Schwartz, 2008, p. 119). So, updates of the code were planned from the beginning, as part of the company's "ethics programme". The code was revised in 2013 -a process documented herein -and released in 2014. After that, and until the moment this case was elaborated, ethical performance management mechanisms were developed and strengthened through the creation of a corporate index of ethical performance. The inclusion of ethics in performance assessment and in welcome training, the ethical risk analysis and the development of a code of conduct for senior management based on the revised code of ethics, were examples of support instruments.

Background
The process leading to the code revision was triggered when preparing the first training sessions on the code of ethics for the employees.
At that time, it was identified that, although the code had been sent for implementation in the different BU, it had not been adopted uniformly in all of them, namely in Brazil and Spain and, to some extent, minor changes had been made to the text so that it would fit in and respond to what were considered, by each geography, to be real needs. This situation according to the ethics ombudsman, was not compatible with what was intended: "that is, the code was to be used as a reference, namely within the ethical performance management system -a document to be seen as a company commitment, facilitating internal and external scrutiny of company's actions". (EO) The search for tuning was to identify the starting points for the code future review.
Another starting point was the change in context. Since the launch of the first code, there had been major changes in the Company's business context: market liberalisation, business diversification, increased involvement of suppliers as a result of increased services outsourcing and, above all, a change in the geographic multi-national scope of the company.)

The Review Process
When the code review began it was decided that it would be carried out with the support of a group of about two dozen people, from Business and Organisational Units (OU), created to support the ethics management (Ethics Focal points). The involvement of this group was, as Martens (2005) considered, the first step in building a global code. The members of the group "are responsible for providing input on content, ensuring that the document addresses relevant issues in each of the major geographical areas where the company operates, and discusses the standards in a way that is meaningful in each culture" (Martens, 2005, p. 1). That was exactly what was expected from the group.
In addition, to support the Ethical focal points' group in the reflection and approval of contents, a restricted team was set up with a more operational function, such as carrying out searches, analysing internal documentation or writing the code itself. The team was composed by the ethics ombudsman, an element of the Sustainability and Environment corporate office, which had participated in the writing of the previous code, and two external elements from the Academia, as consultants, where we were included.
The involvement of those groups, the focal points and the restricted team, makes a difference in relation to the original code-making process, in which only a small number of people were involved: the general secretariat and the sustainability director. In this update, more than 20 people actively participated in the aforementioned review, with a total of about 100 hours of meetings between the two teams.
The focal points group met three times during the code-making process (over 10 months) with an average of 17 participants per session, an aspect to be highlighted given that each session lasted 4 hours and the participants had, in general, high seniority in the company.
How was that possible? According to the Ethics Ombudsman, because: "A climate of construction and trust was created, a climate in which it was shown that the participation of people was important, that they were heard, not as a proforma to say that they had been consulted" (EO).
That is, there was the conviction that "Employees should be involved in the process, not necessarily in order to gain their 'buy-in' or sense of ownership, but to help ensure a relevant and realistic document is produced" (Schwartz, 2004, p. 339) -a document that met the real needs of the Group.
Concerning the generated dynamics, it is also important to mention how the project was led, by the ethics ombudsman, regarding the restricted team. Despite the fact that two consultants were part of the team, there was a real idea of building a team and not the usual delegation, which is common and, as Scharmer (2015) argues, should be avoided: "the greater the complexity of a situation, the more important it is not to outsource: one must stay in touch with the issue as it evolves" (Scharmer, 2015, p. 125) and this is what happened. There was an assumption of the topic complexity and an enthusiastic embrace of the project by the entire team that worked togetherinternal and external elements, sharing knowledge and doubts throughout the process.
The project began with a plenary session of the focal points group, in which the review guidelines were discussed. The intention was not to make a new code, but a revision, in which the original code would be identifiable.
The mission presented to the group was to make a code that corresponded to a level of group commitment to be respected anywhere in the world: "a code of minimums, in the sense that we all agreed that it could not be diminished. A consensual code, a code more axiological than deontological" (EO).
However the code would have to respond to the needs felt by all the Group's BUs. At the kick-off session, content proposals were already presented at the outset, by some business and organisational units. After this meeting, and already within the scope of the restricted team, a hermeneutic work was done regarding the 2005 code. The purpose was to know the true meaning of the words and what they implied, that is, a work was done to decode the code, in the perspective mentioned by Kaptein and Schwartz (2008).

This work involved:
• Identifying, in the original code, references like "obligation of the company", "the company and its employees" or only "the employees" and the meaning in terms of commitment of each one of these references. • Analysing, one by one each of the topics, in a perspective of clarity of the language which, in some cases, originated a new draft, though maintaining the original meaning. • Identifying repeated aspects throughout the code and the interest of grouping some of the topics.
A cross-check was made between the codes of the different BUs, identifying and systematising differences, as well as a benchmarking of codes of the main companies of the sector, to identify new issues, new concerns and aspects to which the code should respond. The extent of this analysis and, above all, the depth of the reflection that emerged from it, was not something predicted from the outset. The research work, analysis and needs assessment led to the questioning of the code structure itself.
This questioning was not easy to accept, especially for those who had been involved in the development of the prior version. There was a kind of resistance, a modesty in touching the text that, in the origin, had been the fruit of much debate and reflection even if only a very restricted team was then involved. This resistance was only surpassed when the objectives of revising the code were defined and explained in a clear and formal way, even if they were already implicit before. The objectives were: a. Clarify aspects that had been identified as less clear from a language point of view. b. Promote uniformity among BUs -obtain consensus on the fundamental standards, without prejudice to adjustments to adapt to the cultural and business specificities of different Bus. c. Refresh the previous version -following the changes of internal and external context that needed to be taken into account.
It was this explanation that, in some way, makes the resistances disappear and allows the shifting from an operational cycle, in which the change is done surgically and reactively, in response mostly to occurrences and new regulation; to a strategic cycle, which starts with an assessement and definition of goals. This strategic cycle starts with stakeholders interviews (in this case, the various business and organisational units, and international organisations) identifies new aspects, accepts questioning the type of code and creates a new code derived from the incorporation of all inputs ( Figure 2).
However, throughout the entire process, there was somehow a spirit of restraint which was no longer in fidelity to the previous code, but resulted, we believe, from a respect for diversity, and an awareness that this restraint was the safest way to make a code that could serve the Group: "The challenge was asking ourselves: what do we consider to be the least affirmation we can have as a group, and that may work, in the individual circumstances of the various BUs, and throughout the various geographies in which we are present?" (EO).
The objectives were formally presented at the second plenary session of the focal points group, where the proposal for a new structure was presented. That new structure retained two topics from the previous code: the "Scope" and the topic related to "Legislation", although moving to a completely different organisation, divided in "Principles of action" and "Commitments to stakeholders".

Synthesis and Comparison of the Two Codes' Elaboration Process
Source: Own elaboration.
The new structure was approved in plenary. The Group was also presented with a proposal for a reformulated code index, containing a comparison with the previous code divided into three categories of topics: "equal", "reformulated" and "new" -having, for each of the amended points, justifications for the change and possible sources. There has always been a concern for transparency in each change, providing these visible registers which allowed identifying each amendment's context.
In the restricted team, and in the focal point sessions, there was a point-to-point debate, already supported by benchmarking summaries, recommendations and suggestions from organisation's various members.
It should be noted that, beside the discussions that took place in the formal sessions with the focal points group, additional contributions from various BUs and OUs came to the team, as a result of their reviews and concerns. It was a question of attitude, a stance of openness and willingness to consider every doubt, every new idea, to try to understand why, and put oneself in others' shoes. This followed what is proposed in the U Theory: a listener of the different actors in which, the voice of judgment, the voice of cynicism and the voice of fear can be silenced, to move forward in an open mind, open heart, and the open will posture . This was essential to prototype a solution to each problem; in this case, to achieve a common understanding, a consensual design, of what the ethical principles of this business group should be.
In addition, it was also made clear that the intended text, this growing consensual text, would not have to be exactly the same in all countries. It was a basis on which changes could be madehowever never reducing obligations and commitments. It was a take on diversity: "When you have a unique instrument for a series of different realities that are cultural realities, business realities and even stakeholders' (in some cases we have shareholders, in other cases not), we have to give these degrees of freedom" (EO).
For exemple for the company's branch with significant operations in the US, the main issue was the donations to political parties, prohibited in several European countries and in the Group's own code of ethics, but which, in the legal context of the US, were allowed, to the point of considering that companies should provide mechanisms so that employees could make financial donations to entities of this nature.
It was decided to maintain, in the "Integrity" topic, the prohibition of the financing political parties by the Company in any part of the world, and to include in the topic "commitment with employees" the possibility for the company to support them in the collection of donations from personal order to political parties.
The objectives of ethics management in the company were also explained in the revised code. This inclusion is proving to be of great importance because it allows building a strategy in a more structured way, to define contents for the training of employees and even, later, to define indicators for the evaluation of the ethical performance of the company.
A distinction was made between "suppliers" and "service providers that are empowered in any way to act on company's behalf", the latter being now subject to the same standards of conduct as company employees. The distinction was made because, for the external public, in many situations, it is not possible to distinguish between those who are company employees and those who are employees of companies that acted on their behalf and which, in many cases, provided services even at customers' homes. This situation, with the increasing externalisation of services, is common to many companies.
Once the code was written and the text stabilised and validated by the focal points, an attempt was made to make it clearer, to rewrite it so that it had a language that any employee could understand. In fact, in addition to concepts that may differ from company to company and can be clarified in a document such as an "explanatory guide" -for example, the concept of familythere are words corresponding to legal concepts, such as nepotism or money laundering, whose meaning may be unknown to some employees and need to be explained.
As was already done in several European countries, in the drafting of public laws and regulations, the use of a simplified language was tried, using a specialised company. The process, however, proved to be too complex and it was not possible to arrive at a satisfactory compromise text. Alternatively, although it did not meet the same objectives, a code mini guide was made, containing a summary of the principles, commitments, and complaints management process.
At the last plenary session of the focal points group dedicated to the code, it was announced the sending of the code for approval to the Executive Board of Directors and the implementation plans were drawn up with all the BUs.
The Code was approved and was transposed to the different BUs, with almost no changes made by the respective Board.

Process Internal Evaluation at The Time of The Code Disclosure
At the time of the code disclosure, in the context of the preparation of the ethics ombudsman annual report, an evaluation was carried out with the participants on how the review process had taken place.
An interview was also carried out with the elements that had been involved in the drafting of the code's first version comparing the two cases, stating that the review "was a very intense and participated process, with a great deal of participation from the team. They had many hours of meetings, deep discussions (...) there was interaction, there was a mental picture in which people participated." This aspect is also referred by the representatives in the group of focal points of the BU present in the different geographies: "the drafting process of the new Code of Ethics, with the brainstorming that took place in all Group BUs, resulted in a consensual document in which the company (...) identifies itself completely, therefore making its adaptation occur with virtually no changes"; "The revision of the Code of Ethics in Brazil was the reaffirmation of the practice already From these testimonies, one may conclude that, the participative way adopted for the revision of the code, was a key factor to obtain a single text adopted by all. In fact, throughout the entire process, there was no imposition and, on the contrary, closeness and openness were shown by the Corporate Centre, to include what each geography considered relevant, to become part of the principles and commitments of the entire Group.

Internal Evaluation of the Code's Effectiveness and Quality, Three Years Later
For the purpose of our research and this article, three years after the release of the new code, we interviewed the ethics ombudsman to revisit the review process, to know what had been done since then, to understand how the document has been evaluated and about the perception of its effectiveness.
In the three years that followed the disclosure of the new Code, all employees were trained online, not only to make the code known, but also to disseminate the entire ethics management process in the company. With minor adaptations, this training was the same in all geographies.
A series of awareness workshops on corporate culture and ethical risks were also organised, involving senior management in most countries. Workshops for middle management were also put in place in the same geographies, essentially reflecting on ethical culture and on the managers' role -now spelled out in the code -in reinforcing this culture. These sessions, both for the upper management and for the middle management, were not replicated in the some BUs according to the margin of autonomy that has to be given to each of them but "Nevertheless, even for reasons of communication, and internal and external perception of coherence, the company would like to be able to say that the same had been done in all geographies." (EO) Another initiative was the training of the service providers' employees acting on behalf of the company or working on its premises and who, in this version of the code, are equivalent to company's employees. This training was designed with the support of the suppliers themselves which, in some way, complements the fact that, in the drafting of the code, they were not consulted: "This is not to make them participate in the writing of the code, but it is rather to help the interpretation of the code so that it could be made in their circumstances and open the company's willingness to listen to them, to speak to them on an equal basis about these subjects" According to the ethics ombudsman, the effectiveness of the code can be measured by the degree of its incorporation in documents, in different parts of the world and in different areas: "the codes of conduct that were developed thereafter, used this code as a base and a reference, and this is an evidence of its effectiveness. The code is a normalizer" (EO). This middle step would allow metrics to verify the code incorporation in practice. Another sign of the effectiveness and level of incorporation of the code at the organisational level is the fact that, even though people did not feel the need for its review, they recognise that, even in a compliance perspective, it was very useful: "This is also a way to measure effectiveness. It is a formal effectiveness that goes further than strict compliance. Our code is viewed positively by national and international auditors" (EO).
In addition, employees' opinion in training evaluation questionnaires, often refer positively to the code and the complaints' management process: "Most of the feedback we have (and there is many), although unstructured, it is positive. There are no negative comments referring the code" (EO).
The code of ethics' regulation states that it should be periodically revised or, at least, its adequacy confirmed. But what are the signs and mechanisms developed by the organisation to detect a lack of adequacy? According to the ethics ombudsman, a signal can be the existence of complaints that, despite being perceived as making sense, do not fall into any point of the code: "When we feel that a claimed subject makes sense, that there is reason on the part of the claimant and it does not fit in the code, one gets the idea that it might be good to add a new subject. On the other hand, when legislation comes out or when someone sees a need for a code of conduct, the adequacy of the code of ethics is analysed in order to respond". (EO) In 2018 the Company was considering a new code revision, precisely by the emergence of new legislation on the prevention of harassment at work. This aspect was already conceptually incorporated, although not referring directly "harassment", and the very inclusion of the word itself raises a problem that had not yet been discussed: "The English word harassment translates the Portuguese word assédio, but I do not know if it coincides with the Portuguese legal concept of assédio: when you translate a word, it can be translated correctly, but it may have another legal framework that influences its reading. For example, [the word] corruption may be well translated and have different meanings in different countries." (EO)

Discussion and Conclusions
We described the process of elaborating the code and its results in relation to the intended objectives. However, we did not want to stop there, and instead also highlight a factor we believe was decisive in achieving the obtained results. It is something that Otto Scharmer (2015) referred to as the blind spot. According to this author, much research has been done on the results and on the processes in general, but we know very little about the "invisible dimension" of leadership (Scharmer, 2015, p. 6) and that invisible dimension refers to the characteristics of the leader, the quality of attention and intention that he/she puts in the process, "the inner place or source at the origin of his actions" (Scharmer, 2015, p In the revision, there was an evolution in the sense of framing the problems of today's world, which is reflected by the inclusion of "non-mandatory", but necessary issues for society, such as the Precautionary Principle, the commitment to promote sustainable lifestyles, or the recognition of minority rights. In this aspect, the company went far beyond what it was "obliged" to do. If we consider the three generations of codes defined by Stohl et al. (2009), there is a shift from an essentially first and second generation codes -compliance with law and regulation and concerns with internal stakeholders -to a code that (keeping the previous dimensions), clearly already addresses the third generation: "The third generation of codes is triggered by globalisation. This generation establishes standards of ethical performance in relation to global stakeholders and the larger world community, transcending traditional organisational boundaries and the limited view that ethical and legal are synonymous" (Calderón et al., 2012, p. 6).
When we speak about leadership, we are referring to, not only the ethics ombudsman who is the formal process leader, but also the restricted team as a whole, who undertook the task of revising the code with a soulful attitude to which, somehow, the members of the extended group adhered.
Perhaps the reference to this aspect may seem partial and "unscientific", but it is not by chance that Otto Scharmer (2015) refers to it as the least studied part. In fact, apart from the statements already made and the participation rate at the meetings, there is little factual evidence. But the scarcity of evidence does not mean that a fact does not exist and should not be referred to.
The leadership, the person, the leader or the leaders, is, according to Scharmer, what makes that a process, even if rigorously repeated, may not have the same results: "Two leaders in the same circumstances doing the same thing can bring about completely different outcomes, depending on the inner place from which each operates. The nature of this inner place in leaders is something of a mystery to us" .
The characteristics of the leader, or leaders, are variables to consider and are limitations to the possibility of replication of this process with the same results.
Another aspect for discussion, given that this code also applies to suppliers, is the real effectiveness and legitimacy of codes for suppliers: Suppliers, which are today often located in developing countries, are given the codes, told to comply with them and informed that they are subject to audits as a part of the contract. Suppliers generally have little power to negotiate how the audits are carried out and are expected to comply with different buyers' demands; with each buyer carrying out their own auditing procedure. (Spence & Bourlakis, 2009;Jiang, 2009) in (Helin & Babri, 2015, p. 41) In fact, if transnational companies have branches in several countries, they certainly have suppliers in many others, and these suppliers usually have several client companies, each asking them to subscribe to their code.
Ensuring that what is written in the code makes sense in the suppliers' country already poses a challenge: "We have a statement from the suppliers' management on how it knows and complies with the code, and this is important as a formal act, but does not guarantee that any one of its employees, who goes to our client's house or who is here at our premises knows about the code. And to think that these people are going to walk with the our code in their pockets, it´s a naive supposition because they simply will do not do so." (EO). The inclusion of suppliers seems to be, in many cases, given this impossibility of control, a question of compliance, which is, in line with what Roque et al defend, the stage of most most companies "with an approach centered in risk management and avoidance of penalties" (Roque et al, 2020, p13) The issue for legal and cultural differences between countries is a subject to be explored and to be seen on a case-by-case basis. In this company's code, and in what respects to the differences between European and US laws regarding donations to political parties, the fact that the issue was openly dealt and discussed with representatives of the BUs in Europe and the US, allowed a satisfactory and consensual solution. This kind of practices can avoid situations like the entrance of ethics in Germany, in which a code of ethics imposed without a prior presentation and discussion with the workers' committee, caused a lot of problems (prohibition of love affairs between employees). That was seen as an interference in their personal life: "the Court asserted that (...) banning such relationships in general violated the personal rights of these individuals to freely develop their personalities and to update their lives according to their own preferences" (Talaulicar, 2009, p. 353).
The aspect of adapting to the culture and law of different countries is one of the most important points of reflection in the elaboration of codes by transnational corporations, which can, as Tan and Wang (2011) refer, adopt one of four strategies: a strategy of "Defiance", ignoring the laws of the host country, which safeguards its reputation in the country of origin but may damage the local reputation; a strategy of "Camouflage", yielding to the laws of "host", as in the case of Google in China: "Google.cn, was blamed for kowtowing to the Chinese government and self-censoring what the government deems inappropriate" (Tan & Wang, 2011, p. 381); a strategy of "Compliance", fully complying with local laws, which is often the case with companies from developing countries operating in developed countries; or a strategy of "Negotiation", which is not limited to accepting or denying law and regulations of the sites of operation, but advances to a "dialectical process, in which distinctive logic components are translated and interpreted, and become integral parts of a new hybrid institutional logic" (Tan & Wang, 2011, p. 383), which was the case of the strategy adopted by the Company in the referred case in the US.
Also, the question of translation did not deserve the attention on the part of the companies and even the Academia. By visiting different companies' websites, often with codes translated into dozens of languages, there is a widespread sensitivity to the fact that a word such as corruption, although well translated, may mean different things under the laws of different countries. In fact, "translation could be more than simply dressing the code in new linguistic attire, it could contribute significantly in embedding the code in the local culture and favor implementation of its guidelines" (Tréguer-Felten, 2017, p. 138 This perception of greater security in the axiological approach is also defended by authors like Talaulicar (2009): "one may conclude that global ethics strategies tend to be more consistent, if they rather use principle based approaches for coding ethical conduct" (Talaulicar, 2009, p. 357). "Too specific codes may evoke conflicts because universalising specific requirements faces greater difficulties and barriers. In contrast to more general principles, specific rules tend to clash with more pronounced local customs, cultural, as well as legal norms" (Talaulicar, 2009, p. 358).
In addition to the security given by the axiological approach, there is also the matter of the level of requirement: when a more normative option is made, the level of commitment may be reduced, for instance: "A code that allows accepting or making offers up to €100 is more or less demanding than the code that places in one's conscience the criterion for acceptance? It is possible to advocate that it is for less because this approach reduces personal responsibility. It is debatable".
(EO)These are questions which answers are not self-evident.
Language is another aspect that presents clear opportunities for improvement in the codes of transnational corporations. It is a difficult consensus between the tone of the rule and that of the principle, between the legal text and colloquial language. It is fundamental to guarantee the means so that, at least the employees (and the suppliers when applicable), have a clear understanding of the code: "Provide Clear Guidance (through examples, questions and answers, bullet points, or other techniques) so that employees understand their responsibilities" (Martens, 2005, p. 11).
The evaluating of the effectiveness of codes is directly related to its objectives. In the studied company it is clear that the code should be the document that establishes the ethical principles and limits as well as the commitments to its stakeholders. But how do you assess whether it is fulfilling that purpose? The implementation of the principles and the respect for the limits will be easier to measure for some themes and countries, but it is a shifting ground, according to Kaptein and Schwartz (2008) in 79 empirical studies on the subject. They reached conclusions in all directions: "35% of the studies have found that codes are effective, 16% have found that the relationship is weak, 33% have found that there is no significant relationship, and 14% have presented mixed results" (Kaptein & Schwartz, 2008, p. 113).
In any case, two factors influence the evaluation: the ambition of the objectives and the language used, especially as referred by (Martens, 2005): "Companies should avoid (…) 'never', 'at any time', or 'under any circumstances' in code standards, except where the company truly means this. In many of the cases, the company may make exceptions and employees will be quick to note inconsistency" (Martens, 2005, p. 11).
An interesting approach that allows an evaluation of authenticity is the one proposed by Kaptein (2015). To present three anonymous codes to a group and to have them guess which belongs to the organization. If people think for example that the code is the one of a company in a wholly different industry or that they think that the code is a multinational whilst it is from a national company (or the opposite). (Kaptein, 2015, p. 58) we don't have a very authentic code.
We believe that a code made in a participatory manner has a better opportunity of being considered authentic. The process also represents an ethical challenge, due to the collective thinking; the need to listen to others and to make a collective reflection and analysis of the company's macro and micro ethical challenges; and the attempt to put it in writing in a way that allows better management and inspires better action Finally, we know that the existence of a code of ethics, despite how good the process for its development, and despite how good the opinion of the parties involved in the process and even the stakeholders, does not guarantee per se an ethical action on the part of the company but it is certainly an important tool to support the development and reinforcement of an ethical culture for companies that want to implement it.