*2.1. Research Context*

We chose management consulting as our research setting because it is well known that consulting managers put a lot of pressure on their employees. Alvesson and Robertson observed, for instance, that consultants frequently work more than 60 h a week [55] (see p. 221). Additionally, Gill found that promotions can only occur through high commitment, so workers constantly feel anxious about their current status and performance [44] (see p. 309). It makes consulting an extremely demanding profession with high levels of stress and burnout [17,20,56]. Society views such high demands and their negative health effects on consultants as "defying morality" [57] (p. 807). In fact, anonymous critiques indicating morally dirty leadership in consulting abound on public Internet forums (see a summary of these critiques in Table 1).


**Table 1.** Moral taint indications of over-demanding managers on Internet forums.

Thus, in the public eye, consulting managers are seen as very demanding in several ways. Such socially constructed dirtiness is also visible in the television series House of Lies, loosely based on a novel by Kihn [58], and we see leadership in consulting criticized in autobiographical accounts of ex-consultants as well [59,60]. These worries are confirmed in several academic studies on consultants' work life [16,17,19,21,44,55].

#### *2.2. Research Design and Sample*

In order to explore moral taint assigned to consulting managers, we performed 24 semi-structured interviews. The interviews were conducted with 12 consultant managers, sometimes also called senior consultants, and one associated junior consultant each. The dyadic design helped to compare interpretations between juniors and managers on leadership experiences and its dirty nature. More than half of the interviewed consultants work at big international firms, the others at consulting firms mainly working for the Dutch market (small- and medium-sized). Specializations are diverse, as indicated in Table 2.


**Table 2.** Interviewee characteristics.


**Table 2.** *Cont.*

\* Number of employees in consulting departments based on http://www.vault.com/. Last accessed: 18 May 2016). Small: <100 employees; Medium: 100–500 employees; Large: >500 employees.

We selected respondents through "convenience and snowball sampling" [61] (p. 127), with the first few interviewees recommending possible candidates at other firms, mostly starting with the juniors and then connecting to their managers. Juniors and managers with the same number work together.

#### *2.3. Interview Procedure*

The interviews lasted an average of 45 min, ranging from 30 to 60 min. We offered anonymity, requested permission to record and started with a short personal introduction of interviewers and interviewees. All interviewees were informed about the study beforehand and gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the guidelines of the School of Business and Economics at Vrije Universiteit Amsterdam. We explained to the interviewees that we would talk about tensions in the manager–employee relationship, based on three jokes (two of them were cartoons, and one was a text joke). The aim was to explore how consultant managers and their juniors experience the dirtiness of the management pressures. The three jokes were a starting point for doing very open interviews, in which we discussed each joke (see Table 3) for 5–15 min. The first one resonated most with the experiences of consultants resulting in long conversations, with the last one being the least.

**Table 3.** Three consultant manager jokes indicating moral taint.


http://www.weitzenegger.de/en/to/jokes.html, accessed: 23 March 2017.

As jokes do not present the truth literally, it helped us to introduce our topic in a stimulating, but when reflecting on it, not a leading, but rather a very open way [62]. Interviewees were first asked to interpret the jokes (for instance junior 9 said, "cartoon 1 is exaggerated."), then they could explain if or how the jokes related to their work contexts ("this (80 h a week) rarely happens here"), and what further associations they had. Most respondents recognized aspects of the dirty management style illustrated in cartoon 1. Related to cartoon 2, the first response of junior 1 was: "What do they mean by this? That you always need to be happy at work? Or that you should pretend you are happy? With that I agree, as you don't want to show your boss you don't feel happy." Many respondents recognized aspects of cartoon 2 due to their own personality or an "up or out" culture in their consultancy. For the same reasons of personality and company culture, others felt less connection to this cartoon. The third joke was hardly representing how respondents felt about what they have to do, and they did not recognize this dirty image of the job, like manager 2, who said, "I would put banker here instead of consultant." All consultants stated they were proud enough to tell what they do, and did not recognize the suggested shame for being a consultant. However, some did refer to other "sick stereotypes" they encountered, like junior 9 who went on holiday, introduced herself as a consultant, getting the question in return, and said, "where is your lease car and credit card?" The quoted interpretations illustrate that respondents made sense of all three jokes in their own way, by referring to their own experiences.

After this free interpretation, the interviewer facilitated a broad discussion including probing questions concerning over-demanding managers, observed critical evaluations of the behaviour of consultant managers, the experienced effects of their leadership style, and how managers and juniors where coping with the situation of pressuring leadership and its morally tainted nature. None of such coping was suggested in the jokes, with only the pressures and the reputation proposed. Starting a conversation with a respondent by asking for interpreting three jokes is new, but doing open explorative interviews aligns with prior research on dirty work (cf. [3,24,33,34]). It is a good way to explore experiences with work pressures, leadership and dirtiness, and it fits our nascent field of research as outlined by Edmondson and McManus [63] (p. 1170).

## *2.4. Data Analysis*

To analyse the interviews, we worked mostly abductive. We applied elements of a grounded theory approach in our coding [64] but also used existing dirty work and moral leadership tactics to interpret the data. We kept an eye out for any codes that did not fit the existing theoretical labels. To do so, the transcribed interviews were coded with the qualitative data analysis tool Atlas.ti (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany). This resulted in 814 relevant codes with data-driven summarizing labels mostly connected to one quote only, and incidentally to two. Both authors coded iteratively and pointed upon which they disagreed with were discussed and then aligned. An overview of all codes can be found below in Table 4. In the results section we present codes related to dirty leadership pressures and related effects in Table 5, codes related to normalization tactics in Table 6 and codes related to moral leadership tactics in Table 7.

The leadership pressures and their effects on juniors were coded as morally dirty based on two ethical perspectives: deontology (pressures) and consequentialism (effects). Criticized stressors, such as long working hours and high work pressure, were coded as morally dirty from a deontological point of view. For example, demanding more hours than allowed by law does not conform to duty [65]. Codes identifying criticized negative effects, such as burnout, decreased wellbeing, or high turnover rates caused by health problems, were coded as immoral from a consequentialist perspective [66]. Normalization and moral leadership tactics were labelled with existing concepts from the discussed literature, except for the new ones that emerged from the data. Our findings suggest that moral taint experienced by consultant managers is not only mitigated by taint normalization, but also by known and lesser known moral leadership tactics. A process model summarizing the main codes and sub codes (in the boxes) and their relationships (arrows) is presented at the start of the results section (see Figure 1).


**Table 4.** Parent, child and grandchild codes.
