“Culture Is What You See When Compliance Is Not in the Room”: Organizational Culture as an Explanatory Factor in Analyzing Recent INGO Scandals

Abstract In recent years, leading international nongovernmental organizations (INGO) such as Oxfam International, Save the Children, and Amnesty International have been implicated in scandals about sexual abuse and other forms of abuse of power and harassment. I suggest focusing on organizational and sectoral culture as an explanatory variable for these crises, which are particularly hard hitting for purportedly value-based organizations. In the case of NGOs, in my observation as a practitioner and ‘pracademic’, these are driven by six factors:(1) particular individual leadership traits that may be prevalent especially in the emergency and humanitarian relief related sector; (2) the effect of power on leaders’ perspectives and behaviors; (3) a culture of silence that makes it hard for NGO staff to speak up about toxic workplace behaviors; (4) the presence of deep power structures within NGOs which are not openly acknowledged and therefore addressed; (5) the myth of own innocence that leads NGOs to treat wrong doing as aberrations instead of systemic problems; and finally (6) a culture of limited individual and team-level accountability practices. The extent to which these cultural failures can be addressed through formal policy and (self)regulatory measures is limited, given that culture is primarily about informal, covert belief systems. NGOs will have to do sustained and disciplined culture work themselves if the roots of the scandals are to be taken away.


Introduction and Setting
In recent years, well known examples in the INGO world including Oxfam Great Britain, Amnesty International and Save the Children have been affected by the sexual abuse crisis. Besides sexual abuse, organizational surveys in these organizations report staff complaints and allegations with regard to 'toxic workplace behaviors', such as bullying and harassment as well as racism and sexism. 1 For NGO practitioners this raises a profound question: what does this crisis tell us about organizational culture? By culture, we refer to the set of shared habits and behaviors (Rock, Grant, and Slaughter 2019), often based on subconscious belief systems and norms ) about how to behave as well as succeed in a certain NGO.
This essay will suggest some cultural factors that may drive these crises and will do so through a decidedly 'pracademic' perspective. 2 The factors point to areas where there seems to be disconnect between the 'espoused' versus 'real, in-practice' behaviors of NGO staff ). They include: (1) specific individual leadership traits that seem to be more prevalent in the emergency and humanitarian relief related sector as compared to other NGO sectors  and that may pose greater risks for sexual abuse; (2) the effect of power on leaders' perspectives and behaviors Piff et al. 2010; Rock, Grant, and Slaughter 2019), which can make leaders less sensitive to other people's perspectives as well as more prone to engage in high risk behaviors. I also discuss a group dynamic level factor (3) a 'culture of silence' that makes it hard for NGO staff to speak up about toxic work place behaviors (NeuroLeadership Institute 2018). Organizational level factors also play a role: (4) the presence of deep power structures within NGOs which are not openly acknowledged and therefore addressed (Batliwala 2011) and which may condone certain behaviors as well as reflect real in-use norms; and (5) the myth of own innocence in 'virtous organizations' (O'Hara and Omer 2013) 3 that may explain instances of deep psychological resistance among NGO staff to the notion that their organizations too can do real harm to the very people they purport to support. Staff as a result may treat wrong doing as aberrations instead of systemic problems, the myth can trigger denial and rejection of self-criticism and can prevent peer monitoring and feedback. Finally, I address (6) a culture of limited individual and team-level accountability practices within NGOs as an example of a factor at the institutional level. The essay will close with some implications for nonprofit policy and regulatory measures, whether within NGOs as individual organizations, for NGO sector level umbrella bodies and self-regulatory schemes, or for government entities. However, given that culture is primarily about informal, covert and subconsciously held belief systems, I also suggest that these types of formal policy and regulatory measures have clear limitations, and I thus point to specific culture work that NGOs will need to do to complement policy efforts.
Let's first look now at an operationalization of the fairly abstract, frequently misunderstood yet easily bandied-about term of 'organizational culture'.

A Definition of What Culture Really Is
For a definition of culture, I draw upon my 2018 essay for CIVICUS' annual report on the State of Civil Society (Vijfeijken 2018) 4 : In this essay, I argue that 'organizational culture' is a concept that, while increasingly used in NGO parlance, is not necessarily understood. It is often mixed up with organizational strategy, for example. I appreciate the NeuroLeadership Institute's definition: culture is the instinctive and repetitive patterns of behaviour guided by deep, shared beliefs about 'what is right' and 'what works'. 5 These belief systems may be influenced by founders or subsequent leaders, but may also be (re)shaped over time through collectives of aligned staff We should understand culture as the DNA, or personality, or mindset of an organization. We can 'see' culture through visible behaviours, which are in face guided by invisible beliefs, values and assumptions about 'the way we do things around here'. In this manner, culture provides guidance -whether intentional or not -on what is done, or not done in an organization; how it is done, and why. 6 It guides staff to what is desired, valued and rewarded in an organization. We are therefore talking about self-sustaining, and quite tenacious patterns of behaviour, that have worked for the organization before and thus have been reinforced over and over again. While culture can be reflected in organisational philosophy statements or other official documents as deliberate attempts by leaders to shape values and behaviours, culture as such primarily works at the level of the unconscious, through informal ground rules or unofficial guidelines of 'how to succeed here' (Katzenbach 2013). 7 In short, culture has overt, but especially covert aspects. Figure 1 maps the factors that I discuss in this essay onto the covert and covert aspects of culture. Let's now look at the first factor which can lead to sexual abuse as well as other toxic workplace behaviours in NGOs: the potential role of individual leadership styles.

Individual Level Factors Individual Leadership Styles in the Humanitarian Sector and Their Role in Cultural Signaling
Much of my research on INGOs in the last 15 years has been grounded in research undertaken by the Transnational NGO Initiative at the Maxwell School at Syracuse University. Our research has been characterized by a strong focus on INGO leadership level perspectives. The basis for this was a National Science Foundation funded interview study (Hermann et al. 2010) 8 with 152 leaders of US registered NGOs, which took place between 2005 and 2008. As part of this focus, Margaret Hermann 9 and Christiane Page analyzed the interview data for differences in individual leadership style. Their question was: do the CEOs of humanitarian and development NGOs exhibit different leadership styles and perceive their work environments in different ways? And if yes, does this indicate a need for match between the organizational context and the individual leadership traits? This research question is based on Fiedler's contingency based leadership theory 10 (Fiedler 1978). Hermann and Pagé 11 found that those in charge of humanitarian NGOs were more likely to challenge the constraints in their environments, and to be interested in influencing what was happening then leaders working in long term development NGOs. Humanitarian NGO leaders tend to be focused on addressing the short-term needs of those in the communities facing the crisis, disaster, or emergency. These leaders viewed themselves as having a short time in which to respond, exhibited a 'take charge' attitude, and they chose to communicate and act informally. In addition, humanitarian sector NGO leaders, style wise, were found to be interested in gaining as much control over their situations as possible, given that they work in settings where their own security and that of others may be at risk. One can reasonably speculate that these 'take charge' leadership traits -while warranted given the often acutely challenging circumstances -also create an environment in which abuse of power, including sexual abuse, is more likely. 12 In addition, the short-term focus of humanitarian sector leaders on saving lives and producing goods and services in as short a time frame as possible can lead to 'savior' complexes and 'macho' or 'cowboy' behaviors. These behaviors are frequently self-reported by practitioners in the humanitarian sector. In addition, aid situations frequently exhibit an inequality in power, between the 'haves' and 'have not', those who have resources to dole out, and those who have to play the role of recipients of those goods. One can speculate with a certain measure of reasonableness that in a humanitarian emergency setting, this inequality of power is even more acute -and that this, in combination with the 'savior' or 'cowboy' self-image noted above, may lead to an environment in which abuse of power is more likely to happen. 13 In sum, while the research findings of Hermann and Pagé on the match between sector and leadership traits may be seen as a structural factor, as a practitioner I argue that sustained and persistent leadership behaviors coming out of the individual leadership traits described send cultural signals to the rest of the organization that certain abuses of power may be condoned, if not sanctioned.
Next, I will discuss how having organizational power as such affects the perspective and way of thinking of individual NGO leaders.

What Power Does to Leaders
Recent neuroscience also offers us insights on what power, neurologically, does to individual leaders. 14 As leaders gain power, they become more goal-focused, very focused on the 'why' of certain actions (as different from being concerned with the 'how'). This further reinforces the fact that leaders are also frequently selected as well as promoted for having a strong goal orientation as it is. Having power makes leaders also lessen their inhibitions around rules and norms and makes them less aware of how their behavior may impact on others. 15 It thus lessens leaders' ability to take the perspective of other people. Equally, having power makes leaders see other people in a more instrumentalist manner, as a 'means' to an end. Similarly, power makes leaders focused on abstract, high level thinking rather than the operational details, realities and the 'how' of everyday organizational life -and how staff may be impacted. If neuroscience related research shows this for private sector organizations, there is ample reason to think that NGO leaders are not precluded from this phenomenon.
Moreover, power, neurologically, makes leaders more optimistic, 16 and lessens their awareness of risks -in fact, it makes leaders more attracted to taking risks. Power increases leaders' sense of certainty. This implies that NGO leaders too (just like in other sectors) are at a higher risk of engaging in abuse of power, including sexual abuse and other forms of harassment and bullying. These leaders simply may be less concerned with how their behavior may affect others, and at the same time they may feel they 'can get away' with it.

A Group Dynamic Level Factor
Power is exercised within groups in the workplace. On a day to day basis, employees of NGOs may simply find it hard to speak up about toxic behavior in their group dynamics. Let's look at why this may be the case.

Toxic Behavior at Work in NGOs, and the 'Grey Zone Behaviors' that May Facilitate It
Toxic behavior at work in any type of organization can consist of bullying, harassment, putting people down etc. While some of the worst forms of these behaviors can lead to formal sanctions such as negative performance records, withholding of promotions, fines, removal from the organization and litigation, there are also shades of 'grey zone behavior' that can create a climate that enables these most toxic of behaviors to take place.
Recent neuroscience as it relates to organizational behavior explains why it can be so hard for employees to speak up in the face of toxic behaviors. The NeuroLeadership Institute's research on this has been illuminating. And there is no reason to assume that NGOs are any different from other types of organization in this regard. According to NLI's research, 17 employee bystander behavior is determined by answers to questions such as: is it ambiguous or not for me to speak up about toxic behavior, or should I just let it go? Am I the person who should speak up as bystander? 18 Second, is there a threat involved in speaking -either a threat to me, or a threat to the people about whom I am about to speak? Third, and most obviously, what is the power dynamic? If the perpetrator of sexual harassment or abuse has more positional power, the threat level increases, due to fear of retaliation or loss of a job.
Moreover, is the culture of the NGO where this occurs comfortable in calling out behavior? Given the weak internal accountability systems and practices which I have observed in my practitioner related work, and which I will turn to towards the end of this essay, there is reason to question this. The problem, according to NLI, is that "silence creates tacit approval" for such behavior. 19 Leaders are obviously important in either creating, condoning or alternatively disrupting such silence. And, according to Broucek at the University of Michigan, such conflict avoidance can lead leaders to "unconsciously allow negative behaviors like bullying to poison the well". 20 NLI asserts that the creation of a "voicing culture" thus is important. 21 In such a culture, staff feel empowered to raise their voice early on; they feel less helpless or conflicted. This culture creates an "environment of consequences" as well as clarity and understanding on all sides -not just on the side of potential victims or survivors, but also on the side of potential perpetrators.
However, in an NGO practitioner task force on culture change 22 which I participated in, we reflected upon a culture of 'niceness' which many of us observe . This can otherwise be described as a culture of conflict avoidance. 23 The 'social contract' -the set of norms on how to behave and otherwise 'fit in' -in those NGOs is that "we are all nice people and we are all doing our best." 24 Parenthetically, it is my contention that it is not coincidental that among the big (con)federated INGOs that have affiliates or members, the term "NGO families" is quite prevalent. I would argue this is a cultural artifact. Seeing the organization as a family suggests a strong 'human resource frame' 25 in which human relations are paramount. Moreover, according to Jennifer Garvey Berger, 26 the notion of the organization as a family runs the risk of the 'mindtrap of agreement', in which conflict aversion is indeed the rule.
Of course, this is a generalization, and thus there are exceptions one should also mention. I note, for example, that the large campaigning NGOs on the whole have a much tougher, and less conflict averse culture. In these NGOs I observe an emphasis on the use of persuasion as form of communication, and harsh language, contentious internal politics and confrontation are common . In these cases, the external campaigning style of communication is mirrored internally. That 'macho culture', though, can stand in the way of true accountability if the behaviours of 'machos' (which can include women) are not called into question.
Let's now turn to organizational level factors that can also seriously constrain the chances that sexual abuse will be exposed.

Politics and Power within INGOs: The Presence of 'Deep Structures' that Enable Sexual Abuse and Other Abuses of Power
It is self-evident that every organization -INGOs included -are also seats of politics and power, in which people jockey for positions, struggle with each other over limited resources, try to get access to top decision makers to increase influence and form (in)formal alliances to gain power. 27 Batliwala (2011) 28 and Moss Kanter (1977) 29 have pointed to the existence of what Batliwala coins"deep structures" in organizations -including in those that aim to do good. Batliwala defines such 'deep structures' as the "sites of invisible and hidden power and influence in organizations". 30 They can consist of informal cliques that wield indirect influence; the actual versus the stated norms; and what kind of values are rewarded in practice. NGOs are not impervious to these power dynamics and in this sense too cannot claim they should be put on a higher moral pedestal. In fact, when such power structures are not acknowledged because they are counter to how NGO staff aspire to see themselves, they are less likely to be addressed. 'Deep structures', when manifest, are a subset of the overall culture of an NGO, focused particularly on sites in the organization where true power and influence is held. They "influence the implicit culture in INGOs, the informal values and systems of influence, reward and recognition" 31 that hold sway in NGOs. And of course, these power structures are evidence of the culture, since they are the products of unconscious belief systems. Some of these belief systems consist of "unspoken, unwritten rules about male and female roles and behaviors, how people are expected to present themselves and what constitutes 'leadership". 32 In my 'pracademic' work I have noticed that certain behaviors which amount to bullying or harassment will be 'overlooked' by some NGOs when the person exhibiting the behavior otherwise is seen as a 'hero' within the organization -whether it is due to charismatic leadership attributes, campaign successes or fundraising prowess (the 'rain maker'). Moreover, in a value-based sector like the NGO sector, where passion for the cause is prized, aggressive ('cowboy') work styles can be misinterpreted as signs of passion for the cause. Clearly, abusive or otherwise toxic behaviors are not as likely to be called out in those instances. This is one more reason why sexual abuse and other forms harassment and bullying may continue without interruption.
Let us now shift to a deeper psychological -and more disturbing -dynamic that may be at play in the case of value-based organizations such as NGOs.

The Dark Side of 'Virtuous Organizations'
How is it possible that organizations that claim to be values-based and wish to promote virtue in the world, sometimes expose an 'underbelly' in which some of their staff, clients, constituents or beneficiaries, are mistreated? According to the psychologists O'Hara and Omer, 33 "virtuous organizations may maintain a myth of their own innocence despite actions, attitudes and values that run counter to this stated view of themselves." The authors' views on this are required reading, in my view, for anybody interested in the organizational culture of INGOs.
According to O'Hara and Omer, activities that affirm the espoused identity, beliefs, attitudes, and values of the INGO are prized, while activities that undermine these assumptions are fiercely resisted. To maintain a sense of innocence as well as internal coherence, INGOs can show strong cultural defenses as well as defensiveness against any allegation that contradicts that myth of their own innocence.
Staff of INGOs can be passionately invested in their mission. I have often reflected that -while according to many people this is a positive factor -, it also has several less positive side effects. One implication is that NGO people tend to want to be involved in decision making on relatively small details in their organization, which can lead to over-involvement and drawn out consultation processes without sufficient closure. I have observed this many a time in the mid to large size global INGOs I have collaborated with or consulted for. This over-identification with NGO work shows up more starkly, though, when the coherence and "authorized and accepted version of the organization's identity" (O'Hara and Omer 2013) is threatened. Disbelief, denial, scapegoating, attacks on leadership and even rage and intimidation of staff involved may result. The extent to which NGO staff's very identity -especially at the leadership level, where my work has been concentrated -is tied up with their work is visible in many ways. When O'Hara and Omer then assert that "such scapegoating and attacks on leadership may create the illusion that problems have been addressed, thereby allowing the myth of own innocence to maintain", 34 it is plausible that this has also happened in NGOs. If your very identity is tied up with 'doing good', then much is at stake. Alternatively, the need to maintain a "collective sense of virtue" 35 is strong because NGO staff's individual identities are so intertwined with the NGO work, in my observation, that organizational cover up practices may be condoned. Not to speak, of course, about the strong organizational tendencies to self-protect out of fear of losing donor funding.
O'Hara and Omer also maintain that "demands on accountability are seen by staff as an unwelcome intrusion into authentic relationship they feel they maintain". Certain of their own individual innocence, members of virtue-driven organizations such as NGOs may consider themselves beyond the need for the kinds of conventional constraints on their beliefs and behavior that less "enlightened" groups require. All together, these are unhealthy and potentially disturbing dynamics in what are supposedly value based organizations, which altogether can help explain why abuse and harassment in the workplace is feasible even in organizations that claim to 'do good'.

What Should We Do in Response to Cultural Failure? Policy and Regulatory Measures, and Their Limitations
Much has been written in the last 5-10 years about the increased need to pay attention to the accountability of INGOs (for standards of behaviour, outcomes and otherwise). Until relatively recently, NGOs did not have to show much accountability other than for overhead ratios and narratives of examples of success -and then primarily towards boards and donors, both forms of 'upward accountability'. 36 As Prakash indicates, 37 though, in comparing NGOs to firms: Conceptually, it is difficult to assert that managerial abuses (agency conflicts) will be less muted in nonprofits versus firms. One plausible argument might be that the cadre of people joining NGOs is intrinsically more 'moral' than the ones working for firms. (Prakash 2007) That particular argument, as I argue elsewhere in this essay, may also have been held onto too tightly by NGOs, and needs to now be challenged.
While the need to be externally accountable in a more systematic way to a greater set of stakeholders has since become more accepted, NGOs internally did not feel strong accountability pressures when it comes to individual behaviors. 38 In my 'pracademic' observation of at least 10-15 mid to large size INGOs, historically, for instance, there has not been a strong discipline of insisting on regular performance reviews, including the use of behavioral competencies.
In general, we must distinguish two forms of accountability. There are formal, 'hard forms of accountability', as expressed in internal performance management reviews, official sanctions, board supervision, selfregulatory sector level standards, formal regulatory expectations by government entities, reporting requirements, etc. These need to be distinguished from organizational 'soft forms of accountability' -culture, behavioural expectations expressed during staff onboarding, in competency frameworks and leadership development programs etc -which are equally important when NGOs want to encourage certain behaviours. Overall, seasoned accountability focused NGO practitioners have contended there is significant room for improvement in balancing these two forms of accountability. 39 Clearly, behavioral and culture failures of the kind outlined in this essay will not be resolved through formal policy, regulatory or governance measures alone, whether by government, institutional donors or NGO self-regulatory agencies. Of the six factors highlighted in this essay, internal accountability practices systems that make it easier for NGO staff to speak up when they observe toxic workplace behavior lend themselves best to (self-)regulatory practices. Of course, governments and NGO umbrella bodies can respond with greater oversight and self-regulatory mechanisms, as the Charity Commission in the UK and NGO umbrella bodies such as BOND (UK) and InterAction (US), for example, demonstrate. In the US context, InterAction member CEOs have signed pledges to address sexual abuse, for instance, and this aspiration is now being supported through sector level as well as organizational level capacity building. Self-regulatory standard setting has inbuilt weaknesses of course, because as Prakash and Gugerty indicate "the excludable nature of these benefits is crucial to prevent free riding". 40 Moreover, when self-regulatory mechanisms mean that NGO members pledge adherence without the use of verification or third-party verification (Prakash and Gugerty 2010), this potential weakness is ever more present. We also must have realistic expectations about board supervision as an effective formal check on real in-use behaviors. Of course, boards can provide oversight so that formal whistle blowing, reporting and complaint mechanism resolution mechanisms are in place and these mechanisms have been adequately resourced. Boards can review trends in the usage of these formal mechanisms and track what percentage of cases gets resolved, the extent to which disciplinary action takes place in severe cases etc. And of course, if boards hold the top leader accountable for negative behavior on their end or that of their senior teams, this sends a significant cultural signal. However, boards are often by necessity quite removedexcept in the case of membership-based movements -and it is hard for them to obtain insights into the daily habits and behaviors that are rewarded or discouraged within the organization -until negative media attention and scandals emerge. Boards can insist, however, on employee engagement surveys and 'pulse checks' of staff motivation and sentiments -and on seeing the full, unvarnished results of these instruments.
In terms of government policy and regulation, in the US, state level Attorneys General can scrutinize NGOs for whether they have enough systems in place. Donors can withhold funding or make it conditional on formal measures being put in place -as has been done in the UK for instance. We can see a gap in standards, so perhaps this is where -in the US context again -the Better Business Bureau-Wise Giving Alliance can step in.
At the individual organizational level, some NGOs have responded with system strengthening by putting safeguarding policies strategy and system measures in place. They have appointed additional safeguarding experts and staff, as well as ombudsmen systems. They have increased staff training and included more specific codes of conduct with respect to safeguarding in staff onboarding procedures. This is laudable and needed but does not necessarily get to all the culture failures. This work needs to be complimented with internal NGO culture work of a long-term, sustained and disciplined nature. This should focus on the encouragement of certain habitual and widely shared, repetitive behaviors and the discouragement of others.
Some of the incentives, described in this essay, have to do with power issues, as well as -foremost -the 'myth of own virtue' or innocence. Culture work to address this can be done in several practical ways. First, invest in leadership development programming focused on the capacity of leaders to nurture strong self-awareness, personal accountability as well as humility. Add to this skill development in how to offer frequent, informal feedback and engage in difficult conversations. Value these skills and self-awareness and self-regulation attributes during performance reviews and in talent management processes. Second, involve authentic informal leaders as allies or 'ambassadors'; they walk the talk and 'show by doing'. Use people who are exemplars, hubs in organizational networks as well as early adopters as 'culture carriers'. Third, intentionally 'hold up' and celebrate everyday behavior you wish to see more of -and do so in a sustained, disciplined and widespread manner. Fourth, call out people who do not, as well as those who display behavior that points to abuse of power. Leaders need to watch out for staff who wield power and authority and who treat others disrespectfully or undermine others in meetings by talking over them or ignoring their contributions. Whether these persons are called out or not is something other staff will keenly watch. Fifth, make skillful use of stories, vignettes and images to signal which behaviors the organization seeks. Leaders can symbolically signal the need for humility by, for instance, deferring to people with lower power in meetings by letting them speak first. Those signs can contribute to cultural change, since staff watch what leaders do, not what they say. Sixth and perhaps most crucially, having alignment among senior leadership team members in terms their own behaviors is critical. This, in turn, calls for strong senior team group process skills, since as research on senior team effectiveness (Edmonson, Roberto, and Watkins 2003) indicates, such skillful group process work needs to mediate for likely asymmetric distribution of interests within teams -over and beyond likely information asymmetries And ultimately, leaders need to be on the lookout for any organizational 'blindness' caused by a collective attitude of 'pseudo innocence', as described by O'Hara and Omer (2013) and take this on publicly in a forceful manner.

Conclusion
In this essay, I have discussed six factors at different levels -individual, group, organizational and institutional -that may be at play in NGO cultures. Some factors interrelate and affect each other. For example, for the individual level factors, the 'take charge' leadership style that may be prompted by humanitarian settings and may be rewarded and therefore reinforced in the humanitarian NGO sector may increase the perception of threat to self and to others when it comes to speaking up about abuse of power. This may explain the choice to remain silent and not speak up, thus reinforcing a 'culture of silence'. Equally, if the 'deep power structures' in INGOs remain unacknowledged and not addressed in the organization, this makes it all the more possible for NGO staff to uphold their 'myth of own innocence'.
Among all of the six factors, this 'myth of own innocence' may be most central in any attempt at culture change when it comes to sexual abuse, harassment and bullying symptoms, because combatting this myth would enable work on the other factors as well. Any attempt at culture change thus may need to start with addressing this myth and the powerful psychological resistance it evokes. Bringing this myth from the unconscious into the conscious of organizations will require sustained and disciplined work of the 'soft kind' described above, over an extensive number of years.
In sum, values are "habitually expressed" 41 through culture. NGOs need to develop a deeper understanding of how culture shows up and develop a broader toolset to encourage and instill certain habits and discourage others. In other words, as acknowledged by Danny Sriskanderajah, Oxfam Great Britain's new CEO, "just because we're working to 'do good' doesn't mean we are exempt from causing harm." 42