The Pareto Principle: Italian Land Deeds, Juran's Vital Few and Sugarscape

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

In 1906 Italian economist Vilfredo Pareto, teaching at the University of Lausanne, published his Cours d'économie politique, observing that about 80 percent of the land in the Kingdom of Italy belonged to 20 percent of the people, a pattern he then found in other countries. In 1941 Joseph M. Juran, a Romanian-born American engineer and quality control consultant, came across this research and applied it to manufacturing, concluding that 80 percent of defects come from 20 percent of causes. Juran named it the Pareto principle and coined the phrase the vital few and the useful many, choosing useful so managers would not dismiss the remaining causes as worthless. The 80/20 split corresponds to a Pareto index of about 1.16, and the two numbers measure different things, so they need not add to 100.

The episode explains that Pareto distributions follow power laws with long tails rather than bell curves. Benoit Mandelbrot noted that above a minimum income, the chance of an income doubling is the same whatever its size, a property called scale invariance, which is why financial models built on bell curves fail in crashes. Microsoft found that fixing the top 20 percent of reported bugs eliminated 80 percent of crashes. Pareto analysis ranks causes by frequency until 80 percent of problems are covered, but the episode warns it should be paired with tools such as the Ishikawa fishbone diagram and fault tree analysis so small problems are not left to grow.

• UN Development Programme data from 1992 showed the richest 20 percent of the world held 82.7 percent of its income.
• Physicist Victor Yakovenko found in IRS data from 1983 to 2001 that the pattern repeats inside the top 1 to 3 percent.
• A 2009 Agency for Healthcare Research and Quality report found 20 percent of patients incurred 80 percent of health care expenses.
• The 80/20 rule is used as a benchmark for superspreading, though infectiousness varies continuously across people.
• Joshua Epstein and Robert Axtell's Sugarscape simulation produced an 80/20 wealth split without programming one in.