When Dan Doctoroff, former Deputy Mayor of NYC, is introduced as a dinner speaker he is often compared to a modern-day Robert Moses, the great builder.

I attended a Goldman conference in NY earlier this year on infrastructure renewal, and Doctoroff gave the lunch keynote address. He spoke a lot about the difficulties posed by interest group and interagency conflict during his experiences trying to bring the Olympics to New York and trying to re-vitalize the city.

About half-way through, he put up a slide with the catchy title "Doctoroff's Law", which he cleverly prefaced by first reviewing Murphy's Law (if things can go wrong, they will) and then Metcalfe's Law (value of a telecom network = n^2):

Here is his law (at least my recollection of it, based on what I had scribbled on my napkin):

Difficulty of Getting Project Approved = 2*a + 4*b + 50*c + infinity if its a stadium for the New York Jets

where:

a = # of city entities involved

b = # of state entities involved

c = whether u have to go thru legislature

Everyone had a good laugh, and despite the big grin on Dan's face, I don't think he intended it entirely as a joke. The way he presented it, I think he genuinely believed in the predictive power of the hypothesis, and wanted to communicate his experience in NY to the private sector audience.

As a researcher, it strikes me that it could be fun to design a study to test Dan's hypothesis and to calibrate the 'weighting factors' on each of the coefficients! ;-)
~~

Subsequent comment on this posting by Dick Scott, Jan. 2, 2007:

The Doctoroff approach is very reminiscent of a wonderful short book by Jeffry Pressman and Aaron Wiildavsky, IMPLEMENTATION: HOW GREAT EXPECTATIONS IN WASHINGTON ARE DASHED IN OAKLAND.1973. They point out that projects are delayed in proportion to the number of independent agencies or groups that must sign off on the plans. Assuming that there is an average of 80% probability of approval for each agency (an optomistic assumption), the probability of success after 70 clearances is .000000125.

Posted by rjorr at December 30, 2007 12:07 PM