High employee turnover hinders a company’s long-term success (Heinz et al. 2017, 2020, Hoffman and Tadelis 2021). Fostering a positive workplace climate is essential to motivate, engage, and retain employees (Gorton and Zentefis 2020, Guiso et al. 2015, Martinez et al. 2015). While this relationship is intuitive and has been documented, dysfunctional workplaces with toxic relational climates are ubiquitous. In large and competitive corporations, relational toxicity and anti-professional leader-subordinate interactions have become the predominant causes of employee burnout and quitting. According to a 2019 Society for Human Resource Management report, 20% of US employees quit their jobs in the last five years due to the toxicity of workplace relationships, and 58% of employees who quit because of a poor workplace climate blame their managers (SHRM 2019). Similar trends emerged in a qualitative survey we conducted in Turkey in 2019 involving white-collar professionals: ‘toxic relations’ and ‘difficult leaders’ emerge as the top challenges white collars face in corporate life.
Toxicity in the workplace is associated with the prevalence of antisocial behaviours such as mobbing, gossiping, and the frequent use of disrespectful and condescending language (Akella and Lewis 2019). When leaders adopt such behaviour, it can quickly become the norm within the workplace. In a recent paper (Alan et al. 2023), we conducted a randomised evaluation of an innovative training programme that aims to eliminate toxic relational dynamics within firms. The training programme was offered to 3,000 white-collar professionals in 20 large corporations in Turkey. These corporations operate in six major industries: energy, chemistry, defence, finance, construction, and textiles. A random half of these 20 corporations received the training first, allowing us to evaluate the causal impact of the programme on the firms’ relational atmosphere.
Measuring workplace climate
We use four measurement tools to characterise the relational atmosphere in a firm:
- Administrative data on employee separations, distinguishing between layoffs and quits.
- Lab-in-the-field experiments to measure the prevalence of pro- and antisocial behaviours, including the tendency to engage in toxic competition, trust and reciprocity amongst department colleagues, and a sense of fairness and generosity towards department colleagues.
- Perceived workplace climate: We designed a survey module to characterise perceived workplace quality and relational atmosphere. We constructed indices of workplace climate covering five dimensions: workplace satisfaction, perception of meritocracy within the firm, collegiality, behavioural norms, and prescriptive norms.
- Social and professional support networks: We elicited the social and professional support networks of the companies by asking all employees to nominate at most three colleagues from whom they receive professional (work-related) support and support in personal matters. These detailed templates allow us to measure the degree of social isolation and connection to the leader and construct department network density and various segregation indices.
A workplace climate improvement programme
Firms were randomly assigned to treatment and control groups, stratified by sector. Firms in the control group only had access to the training programme after the endline data collection, i.e. we used a phase-in design.
To implement the training programme, we partnered with a consulting firm established by ex-corporate professionals who aspire to eliminate toxic relational atmosphere and employee burnout among white-collar professionals. The content of the training programme focuses on the benefits of prosociality in the workplace and the importance of professional communication using respectful language, especially in leader-subordinate interactions. Therefore, while the training programme was open to all employees, leaders were particularly encouraged to participate.
The training programme had two phases: online workshops and a project development phase. The first phase involved a series of online workshops emphasising: (i) respectful and peaceful communication with colleagues, subordinates, and leaders, by exerting deliberate effort to eliminate toxic and condescending language; (ii) understanding the others’ points of view and tolerating differences in opinions; and (iii) learning to rely on colleagues by accepting vulnerability (Alan et al. 2023). The training sessions incorporated innovative tools such as creative drama and role-playing and were highly interactive and engaging.
In the second phase of the programme, participants were asked to develop projects aiming to improve communications and relational culture within their firms or to embed these themes in their ongoing projects. After 8 weeks of teamwork monitored by the implementing partner, participating teams presented their projects to the upper management and put them into practice if seen fit. Figure 1 shows snapshots from an actual training session. Due to the outbreak of the COVID-19 pandemic, the training sessions were directed online by the implementing partner. The employee participation, however, was generally in person.
Figure 1 Training sessions on good leadership practices and relying on colleagues
Source: Alan et al. (2023). Shared with the participants’ permission.
Improving the workplace climate
Comparing the participants in the treated and control firms, we document significant positive impacts of the training programme on most of the outcomes we consider.
We find that the employee separation rate is lower, especially at the leadership level, in the treated firms. The likelihood of employee separation is 2 percentage points lower in the treated companies for the overall sample, where control companies have a mean 5% separation rate. Separation of leaders is 4.8 percentage points lower in the treated companies, where control group leaders have a 5.7% separation rate.
Professionals in the treatment group exhibit lower antisocial tendencies in the workplace: they are less inclined to engage in toxic competition and reciprocate their colleagues’ trust more generously than those in the control firms. Figure 2 compares the degree of toxic competition and reciprocity in treatment and control firms.
Figure 2 Pro- and antisocial behaviour in the treatment and control firms
Notes: This figure reports the unconditional means of our measures of toxic competition and reciprocity for the treatment and control firms. P-values associated with the estimated treatment effects are provided. Toxic competition is measured as the fraction of sabotage endowment used in an incentivised sabotage game; reciprocity is defined as the average fraction of the amount reciprocated in an incentivised trust game.
Positive effects on prosocial behaviour are matched by improvements in workplace satisfaction, perceived meritocracy in the firm, and perceived collegiality of the department. However, the positive effects on workplace climate perceptions are limited to the subordinates. A summary of the estimated treatment effects for the subordinate sample is depicted in Figure 3.
Figure 3 Treatment effects on perceived workplace climate and leadership quality
Notes: This figure documents the estimated treatment effects on subordinates’ perceptions of workplace climate and of leadership quality. Values are standard deviation effects, where control group is normalised to zero. P-values associated with the estimated treatment effects are provided.
These findings are complemented by denser and less segregated social support networks in treated firms. A high proportion of employees (13%) in the control firms report that they lack professional support from their colleagues and 24% report a lack of personal support. The training programme lowers the proportion of employees lacking professional support to 8% and personal support to 17% in the treated firms.
A healthy workplace climate requires professional and empathetic leaders
Managers play a pivotal role in setting the tone of communication and thus shaping the relational culture in the workplace (Bloom et al. 2011). Leaders can cultivate a collegial atmosphere or, conversely, propagate a toxic culture of mistreatment and condescending language, reflecting employee behaviour and perceptions (Hoffman and Tadelis 2021, Bloom et al. 2013, Heinz et al. 2017).
We show that the training programme’s positive impact is largely driven by its success in improving leaders’ attitudes toward subordinates. Treated subordinates report higher-quality leadership in their firms. Specifically, they perceive their leaders to be more professional and more empathetic, as documented in Figure 3. Consistent with this finding, treated subordinates are more likely to consider their leaders as primary providers of professional support.
When toxic behaviour becomes the norm, transformative actions may be required to improve the workplace climate and retain employees. Our study demonstrates that an innovative training programme aimed at improving the relational environment can effectively reduce toxic competition, increase prosocial acts, strengthen social support networks, and lower employee separation rates in large and competitive corporations.
Our study underscores the critical role of prosociality in everyday interactions and professional leader-subordinate relations in achieving a healthy workplace climate in large corporations. The toxicity of the relational environment, however, is not unique to the corporate world. It also plagues the public sector and academia. Our findings, therefore, have practical implications for different workplace environments and provide insights into innovative ways to improve such environments.
Akella, D, and V J Lewis (2019), “The modern face of workplace incivility”, Organization Management Journal 16(2): 55–60.
Alan, S, G Corekcioglu, and M Sutter (2023), “Improving workplace climate in large corporations: A clustered randomized intervention”, Quarterly Journal of Economics 138(1): 151–203.
Bloom, N, D McKenzie, A Mahajan, and J Roberts (2011), “‘The Office’ goes to India: Why bad management is keeping India poor“, VoxEU.org, 13 April.
Bloom, N, B Eifert, A Mahajan, D McKenzie, and J Roberts (2013), “Does management matter? Evidence from India”, Quarterly Journal of Economics 128(1): 1–51.
Gorton, G, and A Zentefis (2020), “Corporate culture as a theory of the firm: The role of values, customs, and norms”, VoxEU.org, 16 July.
Guiso, L, P Sapienza, and L Zingales (2015), “The value of corporate culture”, Journal of Financial Economics 117(1): 60–76.
Heinz, M, S Jeworrek, V Mertins, H Schumacher, and M Sutter (2017), “Indirect effects of unfair employer behaviour on workplace performance”, VoxEU.org, 15 December.
Heinz, M, S Jeworrek, V Mertins, H Schumacher, and M Sutter (2020), “Measuring the indirect effects of adverse employer behaviour on worker productivity: A field experiment”, Economic Journal 130(632): 2546–68.
Hoffman, M, and S Tadelis (2021), “People management skills, employee attrition, and manager rewards: An empirical analysis”, Journal of Political Economy 129(1): 243–85.
Martinez, E A, N Beaulieu, R Gibbons, P Pronovost, and T Wang (2015), “Organizational culture and performance”, American Economic Review 105(5): 331–35.
SHRM – Society for Human Resource Management (2019), The high cost of a toxic workplace culture: How culture impacts the workforce and the bottom line, Society for Human Resource Management Research Report.