Week 11: The rat race

Saturation, the rat race modelled and measured in two law firms, the simulation in which a forced distribution's gains fade, and what a room falls back on when nothing separates the names.

18 May 2027

Why it worksWhen promotion runs on a signal everyone can produce, everyone produces it, nobody is distinguished by it, and the race is the equilibrium; modelled, measured in two law firms, and fading within years.

Take a team where everyone has done this course. Every ticket is split, every update is public, every launch has an author, every self-review has a number in its first line, every green dot is on. None of those things now separates anyone from anyone else. The curve still has to be filled. So the room falls back on what is left: who spoke first, who the chair knew, who was newest. The tricks did not stop working; they became the entry fee.

This is not folk theory. In 1996 Landers, Rebitzer and Taylor modelled it and then measured it in two large law firms. Partners cannot see who will work hard once made partner, so they promote on the one thing they can see, billable hours. Every associate knows this, so every associate bills long hours, so hours stop distinguishing anyone, so the bar rises. The associates in the study said, when asked, that they would take a pay cut for shorter hours if everyone else did too. Nobody could move first. That is the rat race: not a failure of the system but its equilibrium, and the paper’s title, “Rat race redux”, says how old the idea was even then.

The forced distribution has its own version. Scullen, Bergey and Aiman-Smith simulated a firm that cuts its bottom ten percent every year and found that the gain in workforce quality is real and front-loaded: most of it arrives in the first few years, after which the easy cases are gone and the curve is being imposed on people who are all roughly fine. Ordóñez and colleagues’ “Goals gone wild” adds what narrow targets do to the people under them in the meantime. Put the three together and you have the whole of this block: the signals saturate, the curve keeps cutting, and the room picks by whatever is left.

The stand-up this week runs the course’s own calibration on you, so you can watch it happen.

Outline#

• saturation: what the tricks look like when everyone has them
• Landers, Rebitzer and Taylor: the model, the two firms, and the associates who could not move first
• Scullen’s simulation: why the curve’s gains fade
• what a room falls back on when nothing separates the names, and how to leave before you are re-baselined

Sources

• Landers, Rebitzer and Taylor, Rat race redux (American Economic Review, 1996)
• Scullen, Bergey and Aiman-Smith, Forced distribution rating systems and the improvement of workforce potential (Personnel Psychology, 2005)
• Ordóñez, Schweitzer, Galinsky and Bazerman, Goals gone wild (Harvard Business School, 2009)
• Don't rate your employees on a curve (Harvard Business Review, 2013)