Coefficient of Inefficiency

Coefficient of Inefficiency is a semi-humorous attempt of C. Northcote Parkinson to define the size of a committee or other decision-making body at which it becomes completely inefficient.

In the book "Parkinson's Law: The Pursuit of Progress", (London, John Murray, 1958) one of the chapters is devoted to the basic question of comitology: how committees, government cabinets, and other such bodies are created and eventually grow irrelevant (or are initially designed as such).

Empirical evidence is drawn from government cabinets from history and contemporary evidence. Most often, the minimal size of a state's most powerful and prestigious body is five members. From English history, Parkinson notes a number of bodies that lost power as they grew:
* The first cabinet was the Council of the Crown, now the House of Lords, which grew from an unknown initial number of members, to 29, to 50 before 1600, by which time it had lost much of its power.
* A new body was appointed in 1257, the "Lords of the King's Council", numbering fewer than 10. The body grew, and eventually ceased to meet when it numbered 172 members.
* The third incarnation of the English cabinet was the Privy Council, initially also numbering fewer than 10 members, rising to 47 in 1679.
* In 1615, the Privy Council lost power to the Cabinet Council, initially with 8 members, rising to 20 by 1725.
* Around 1740, the Cabinet Council was superseded by an inner group, called the Cabinet, initially with 5 members.At the time of Parkinson's study (the 1950s), the Cabinet was still the official governing body. Parkinson observed that, from 1939 on, there was an effort to save the Cabinet as an institution. The membership had been fluctuating from a high of 23 members in 1939, down to 18 in 1954. A detailed mathematical expression is proposed by Parkinson for the Coefficient of Inefficiency, featuring many possible influences. The formula remains to be empirically confirmed. Parkinson's conjecture that membership exceeding a number "between 19.9 and 22.4" makes a committee manifestly inefficient seems well justified by the evidence proposed. Less certain is the optimal number of members, which must lie between three (a logical minimum) and 20. That it may be eight seems both justified and ruled out by observation: no contemporary government in Parkinson's data set had eight members, and only the unfortunate king Charles I of England had a Committee of State with that membership.

References

* "Parkinson's Law, or The Pursuit of Progress", C. Northcote Parkinson, 1957. [http://lib.novgorod.net/DPEOPLE/PARKINSON/parkinson.txt]