500
S&P 500 companies
Topic 01 / Executive Leadership & Demographic Equity
Glass Cliff in S&P 500 CEO Transitions
Do diverse leaders inherit more precarious CEO positions—and what happens after they take over?
The glass cliff describes the possibility that women and other underrepresented leaders are more likely to be elevated to leadership positions when an organization is already facing unusually difficult or precarious conditions.
This study examines the glass-cliff hypothesis across a dataset of 500 S&P 500 companies.
Research summary
Scope of the Topic 01 analysis
Sample sizes are reported at each analytical stage so readers can judge how much evidence stands behind each result.
4
core research questions
433
transitions with measurable pre-handover stock performance
44
pressured/forced transitions with complete post-handover data
Overall Finding
The dataset does not provide broad descriptive evidence of a systematic S&P 500-wide glass-cliff effect.
Diverse leaders were not more likely to receive crisis appointments or follow pressured CEO departures, and they did not systematically perform worse after entering difficult situations.
However, individual extreme transitions show that glass-cliff-type conditions can still occur at particular companies.
Definitions & Methodology
How each measure is constructed
Every classification used on this page follows the definitions published in the Topic 01 report.
- Diverse leader
- An incoming CEO who is female and/or non-White, including diverse co-CEO teams.
- Stock performance vs. S&P 500
- The company's stock return minus the S&P 500 return during the same period. A value of −72.3 percentage points does NOT necessarily mean the stock itself fell 72.3%. It means the company performed 72.3 percentage points worse than the S&P 500 during the same period.
- Crisis appointment
- A CEO transition where the company underperformed the S&P 500 during the 12 months before the new CEO took over.
- Pressured or forced departure
- A predecessor departure involving removal, termination, documented performance pressure, investigation, misconduct, board pressure, shareholder pressure, investor pressure, governance controversy, or another comparable involuntary leadership change.
- Outliers
- Extreme observations identified from the relevant comparison distribution.
Reading the performance measure. A value of −72.3 percentage points does not necessarily mean the stock itself fell 72.3%. It means the company performed 72.3 percentage points worse than the S&P 500 during the same period.
01 / Appointment Conditions
Were Diverse Leaders More Likely to Be Appointed During a Crisis?
| Incoming CEO | Measurable transitions | Appointments after underperformance | Rate |
|---|---|---|---|
| Diverse leaders | 92 | 50 | 54.3% |
| Non-diverse leaders | 341 | 193 | 56.6% |
| Total | 433 | 243 | 56.1% |
Key difference
−2.3 percentage points
What it means
The diverse-leader crisis appointment rate was 2.3 percentage points lower.
Interpretation
The aggregate results do not show that diverse leaders were more likely to receive crisis appointments. However, individual diverse leaders still sometimes inherited exceptionally distressed companies.
Most severe pre-handover outliers
- Dollar General−78.1 pp
- United Airlines−73.7 pp
- Lululemon−72.3 pp
Most severe formal negative outlier among diverse-leader crisis appointments.
- GE−64.8 pp
02 / Leadership Instability
Were Diverse Leaders More Likely to Be Appointed After External Pressure or Forced CEO Exits?
| Incoming CEO | Classifiable departures | Pressured / forced exits | Rate |
|---|---|---|---|
| Diverse leaders | 93 | 9 | 9.7% |
| Non-diverse leaders | 352 | 39 | 11.1% |
| Total | 445 | 48 | 10.8% |
Key difference
−1.4 percentage points
What it means
The diverse-leader rate was 1.4 percentage points lower.
Interpretation
The second question again provides no evidence that diverse CEOs disproportionately enter after involuntary or highly unstable leadership exits. It also demonstrates why the specific circumstances behind a transition matter: two CEO replacements can carry the same label while reflecting very different levels of organizational distress.
Important cases
- Dollar General−78.1 pp
pre-handover performance
Board-led transition combined with severe pre-handover underperformance — the incoming leader inherited both deteriorating performance and predecessor instability. Todd Vasos is a non-diverse CEO.
- Federal Realty+54.7 pp
pre-handover performance
Classification review recommended
The report indicates that this observation may require additional classification review; the company had previously announced Donald Wood's succession as part of Steven Guttman's planned retirement. Its current forced/performance-related classification should not be presented as unquestionably correct.
03 / Post-Appointment Performance
Did Diverse Leaders Perform Worse After Taking Over During a Crisis?
| Incoming CEO | Complete observations | Mean +12M | Median +12M |
|---|---|---|---|
| Diverse leaders | 47 | +3.7 pp | +3.4 pp |
| Non-diverse leaders | 176 | +4.6 pp | -1.6 pp |
Mean difference
0.9 pp in favor of non-diverse leaders
Median difference
5.0 pp in favor of diverse leaders
Total complete observations
223
Interpretation
There is no evidence that diverse leaders systematically performed worse after crisis appointments. The average difference is small, while median performance favors diverse leaders.
Why Mean and Median Differ
Several extremely large positive observations in the non-diverse group raise the arithmetic mean even though the median is negative. This is why both mean and median should be shown.
Major positive outliers
- MGM Resorts+250.6 pp
- Vertiv+210.8 pp
- Caterpillar+157.4 pp
- Devon Energy+156.6 pp
- Albemarle+104.5 pp
- Boston Scientific+100.8 pp
- Carnival+90.6 pp
- United Airlines+81.2 pp
- Ford+80.9 pp
Fiserv−91.6 pp
Independent verification recommended
04 / Performance After Forced Transitions
Did Diverse Leaders Perform Worse After Taking Over Following External Pressure or Forced CEO Exits?
Small-sample caution
Only seven diverse-leader observations satisfy all necessary conditions. The apparent difference should therefore NOT be interpreted as evidence that diverse leaders systematically outperform.
| Incoming CEO | Complete observations | Mean +12M | Median +12M |
|---|---|---|---|
| Diverse leaders | 7 | +18.9 pp | +21.5 pp |
| Non-diverse leaders | 37 | +1.7 pp | -4.3 pp |
Mean difference
+17.2 pp in favor of diverse leaders
Median difference
+25.8 pp in favor of diverse leaders
Total
44
Interpretation
The available data provides no evidence that diverse leaders performed worse after pressured or forced CEO transitions. However, the small diverse sample means the magnitude of the apparent advantage should be interpreted cautiously.
Influential observations
Diverse
- Waters+68.0 pp
- NetApp−20.3 pp
Non-diverse
- NRG Energy+75.2 pp
- Dollar General−51.1 pp
Removing Waters reduces the diverse-group average from approximately +18.9 pp to +10.8 pp.
Open data
Explore the Underlying Data
Review the company-level observations behind the Glass Cliff analysis.
Numeric filters only match observations with a measurable value; rows where the workbook records an explanatory status (for example “+12-month window incomplete” or “no applicable handover”) are preserved as text and are not converted to zero.
| Loading dataset… | ||||||||||||||||
—
MiraeBiz Research Report / Topic 01
Glass Cliff in S&P 500 CEO Transitions
A full analysis of appointment conditions, predecessor departures, post-handover performance, company-level outliers, methodological limitations, and the implications of the Glass Cliff hypothesis across the S&P 500.
- Dataset
- S&P 500 CEO Transitions
- Companies
- 500
- Research Area
- Executive Leadership & Demographic Equity
- Year
- 2026
Methodology & Limitations
What this analysis can and cannot show
The limitations published in the Topic 01 report remain visible alongside the results.
- This analysis is descriptive rather than causal.
- It does not control for sector.
- It does not control for company size.
- It does not control for CEO experience.
- It does not control for macroeconomic conditions.
- Relative stock performance can represent very different underlying types of crises.
- Sample sizes become smaller in the more specific analyses.
Sample sizes narrow at each stage
- 433transitions have measurable pre-handover performance
- 44pressured/forced transitions have complete post-handover data
- 7of those involve diverse incoming leaders in Visualization 4
Conclusion
What Does the Evidence Show?
There is no systematic S&P 500-wide difference in the likelihood that diverse and non-diverse leaders receive difficult appointments. However, the company-level data contains individual extreme transitions that may still reflect the conditions associated with the glass-cliff mechanism.
The aggregate analysis therefore does not establish a broad S&P 500 glass-cliff effect.
At the same time, individual cases such as Lululemon demonstrate why the severity and context of individual appointments remain important.