Where it sits
The entire dispute is about which of these two arrows points which way.
Tell it apart
| Aspect | Glass cliff | Glass ceiling |
|---|---|---|
| You get the job | Yes | No |
| Visible as discrimination | Barely | Somewhat |
| Blame lands on | The appointee | Nobody |
| Named | 2005 | 1978 |
The ceiling stops the promotion. The cliff grants it on doomed terms. The cliff is the crueller mechanism precisely because it looks like a win — and because failure then gets recorded against the person rather than the position.
What it doesn't mean
It was a direct rebuttal to one named newspaper article of 11 November 2003 — and Ryan and Haslam used THE SAME DATASET to show that poor performance PRECEDED the appointments rather than followed them. The causation ran the other way.
A 2009 study — Adams, Gupta and Leeth, in the British Journal of Management — found female CEO appointments preceding BETTER financial performance. The literature records both supporting replications across law, politics and corporate sectors and this contradicting result.
What it really means
This concept exists because of one newspaper article, and the story of how it was answered is the best argument for reading data twice.
On 11 November 2003, a piece appeared reporting that FTSE 100 companies with women on their boards had performed worse than those without — and concluding, in terms, that women were wreaking havoc on British companies.
Michelle Ryan and Alex Haslam, at the University of Exeter, went back to the same dataset.
And found that the poor performance came first. The companies were already failing when the women were appointed. The correlation was entirely real. The arrow pointed the other way.
They published in 2005 and called the pattern the glass cliff: women and minority candidates are disproportionately appointed to leadership roles that are already in trouble — so the position comes with a fall already built into it.
What makes this worse than a ceiling, and the reason the name was chosen so pointedly, is that it looks like a win. A glass ceiling is a promotion you don't get, and everyone involved can at least see the shape of what happened. A glass cliff is a promotion you do get — announced, congratulated, photographed — and when the outcome arrives, the failure attaches to you rather than to the position.
Which then feeds the original article's conclusion. The mechanism is self-confirming: appoint people to unwinnable jobs, observe them not winning, cite the observation.
The evidence is strong rather than unanimous. Replications exist across law, politics and corporate leadership; a 2009 study by Adams, Gupta and Leeth found female CEO appointments preceding better performance. That is what a live, contested and genuinely useful finding looks like — and it is a great deal more than the 2003 article had.
Where it comes from
invisible — borrowed from 'glass ceiling'
a precarious edge
Deliberately built on 'glass ceiling' to argue that breaking through the ceiling can land you somewhere worse than beneath it.
- 11 November 2003
A newspaper article reports that FTSE 100 firms with women on their boards performed worse, and concludes women were wreaking havoc on UK companies.
- 2005
Michelle K. Ryan and S. Alexander Haslam at the University of Exeter publish 'The Glass Cliff', re-analysing the same data and reversing the causal arrow.
In real life
- Ryan and Haslam's finding: FTSE 100 companies that appointed women to their boards were more likely than others to have been doing badly in the months before the appointments, not because of them.
- She was finally offered the chief executive's job — eight weeks after a profit warning, with the board divided and the share price halved.
- 'They gave her the region nobody else would touch, then blamed her when it didn't turn round. Textbook glass cliff.'
Questions
Q01Who coined it?
Michelle Ryan and Alex Haslam at the University of Exeter, in a 2005 paper in the British Journal of Management.
Q02What triggered the research?
A November 2003 newspaper article claiming that women on boards damaged FTSE 100 performance. Ryan and Haslam went back to the same data.
Q03Does the evidence hold up?
Largely, with replications across law, politics and corporate sectors — though a 2009 study — Adams, Gupta and Leeth, in the British Journal of Management — found the opposite pattern. It is a well-supported finding rather than a settled law.
Sources
- Glass cliff — Wikipedia.
- Discovering the glass cliff: Insights into subtle gender discrimination in the workplace — REF Impact Case Study, University of Exeter.
- glass cliff (etymology) — Wiktionary.