The 30–40% finding

Felps, Mitchell and Byington’s 2006 research placed trained actors into teams of four to five people, scripting them to enact one of three toxic behaviours: withholding effort, expressing persistent negativity, or violating interpersonal norms. The results were unambiguous across 40 groups. A single bad apple reduced team performance by 30 to 40%. The experimental design used actors following scripts — eliminating the self-selection confounds that contaminate observational research — making this a clean causal estimate rather than a correlational association. One person. 30 to 40 percent.

The research identified three specific archetypes, each operating through a different mechanism. The jerk triggers perceptions of inequity and interpersonal threat — activating the SCARF status and fairness violations that previous articles in this series established degrade cognitive function. The slacker triggers free-rider perceptions and motivation withdrawal across the team, as equity theory predicts that members reduce their own contribution when one person consistently under-contributes while receiving equivalent rewards. The depressive triggers emotional contagion — the negativity spreads through the team’s collective emotional state through the same mirroring mechanisms established in the leadership neuroscience article, reducing the collaborative energy that performance requires.

Why bad travels further than good

The asymmetric power of the bad apple is not incidental — it is a direct expression of the negativity bias established earlier in this series. Baumeister and colleagues’ synthesis established that bad is stronger than good across psychological domains: negative events produce larger, more enduring, and more immediate cognitive and behavioural effects than comparably scaled positive events.

In a team context, one person’s persistent negative behaviour is processed as a threat by every other team member’s threat circuitry simultaneously, shifting the group from reward-state to threat-state processing. The implication is asymmetric in a commercially important direction: a single positive deviant — someone who brings exceptional energy, generosity, or creativity — cannot produce an equivalent performance uplift to the damage one negative deviant produces. The mathematics of team performance is not symmetric around zero. Negative influence travels faster, penetrates deeper, and requires substantially more effort to counteract than positive influence requires to establish.

Three defensive reactions that all degrade performance

Felps et al. documented how team members respond to a bad apple, and found that every available response consumes resources previously directed at the work. Members who attempt motivational intervention — expressing concerns and asking the person to change — spend cognitive and emotional energy on conflict management rather than output. Members who attempt rejection or isolation produce relationship conflict and political dynamics that degrade psychological safety for everyone, not just the targeted individual. Members who withdraw defensively protect their own wellbeing while reducing their engagement with the team.

None of the three preserves performance. The bad apple does not simply reduce the direct output of one team member — they redirect the cognitive and emotional resources of every other team member away from productive work and toward managing the negative member’s behaviour. The 30 to 40 percent reduction reflects this systemic resource diversion, not just the subtraction of one person’s contribution.

The cost asymmetry that most entrepreneurs get backwards

Housman and Minor’s 2015 Harvard Business School research established a specific and practically important cost comparison. A superstar employee — in the top 1% of performance — adds approximately $5,300 in value over a given period. Avoiding a toxic employee saves approximately twice that amount. The single most valuable hiring decision is therefore not acquiring a high performer. It is avoiding a toxic one.

The talent acquisition effort that goes into recruiting exceptional people should at minimum equal the due diligence effort that goes into screening for toxicity, because the asymmetric return on avoiding the bad apple is greater than the return on finding the superstar. Most entrepreneur hiring processes are calibrated in the opposite direction.

The boundary condition worth understanding

The bad apple research has one counterintuitive limit. Research found that persistently negative individuals can actually improve group performance when the task involves creativity or information processing — because a dissenting voice prevents groupthink, forces more rigorous examination of assumptions, and slows premature consensus. This boundary condition matters for any entrepreneur who might otherwise interpret this article as a prescription for removing all friction from a team.

The distinction maps directly onto the cognitive conflict versus affective conflict framework established in the previous article. The intellectually demanding team member who challenges ideas vigorously is not a bad apple. The chronically negative team member who undermines trust, withdraws effort, or persistently violates interpersonal norms is. Protecting the former while removing the latter is the practical task — and the research is clear that most organisations are poor at making this distinction, particularly when the bad apple has seniority, tenure, or specialist knowledge that makes removal feel costly.

The cost of keeping them is reliably higher than the cost of losing them. The performance degradation is diffuse and invisible in output data; the cost of removal is concentrated and visible. This asymmetry of perception is why bad apples survive — not because organisations genuinely believe the trade-off is worth it, but because the full cost is never made visible until the removal happens and the team’s performance recovers.

A book worth reading alongside this

The No Asshole Rule by Robert Sutton is the most widely read applied treatment of the bad apple phenomenon and the most honest available reckoning with its commercial costs. Sutton’s total cost of assholes framework — which attempts to quantify the full financial impact of one persistently hostile team member across turnover, reduced productivity, management time, and psychological damage — extends the Felps et al. research into a practical business case that is difficult to dismiss. His documentation of why bad apples survive in organisations despite their obvious costs — seniority, specialist knowledge, fear of conflict, ambiguity about what constitutes toxic behaviour — is the most useful available treatment of the gap between what organisations know and what they do. For any entrepreneur who has been aware of a team toxicity problem and delayed addressing it, this book provides both the evidence for why delay is costly and the framework for acting on it.

Have questions about this article?

If any part of this article raised questions you want to explore further, courbot.co is built for exactly that. It is courben.co’s AI assistant, designed around the psychology of entrepreneurship. Ask it anything from this article.

This article is for educational and informational purposes only.

Sources: Felps, W., Mitchell, T.R. & Byington, E. (2006), Research in Organizational Behavior, 27, 175–222 —| Baumeister, R.F. et al. (2001), Review of General Psychology, 5(4), 323–370 —| Housman, M. & Minor, D. (2015), Toxic Workers, Harvard Business School Working Paper 16-057 —| Adams, J.S. (1965), Advances in Experimental Social Psychology, 2, 267–299 —