My theory on Nobel Prizes is that they are awarded for what baseball analyst Bill James calls "peak value" rather than "career value." Sandy Koufax had great peak value, winning 27, 26, and 25 games in his three best seasons, but his career wins were only 165. Don Sutton, Koufax's teammate in 1966, had great career value, with 324 wins, but he won only 21, 19, and 19 games in his three best seasons.
In economics, high "peak value" is represented by one or two highly-cited papers. Myron Scholes, John Nash, Michael Spence, and Ronald Coase exemplify that. On the other hand, folks like Samuelson, Modigliani, and Friedman represend both high peak value and high career value, in that their most-cited works are impressive but are far from their sole claim to fame.
I can think of a lot of economists with high career value but only modest peak value, and they seem unlikely to win Nobels--someone like Robert Hall does not even generate speculation. Very few Nobel Laureates fit the pattern of high career value and modest peak value--the only one who comes to mind is Douglass North. Perhaps George Stigler and Amartya Sen.
Phelps' peak value comes primarily from the "Phelps volume," which propelled the natural rate hypothesis to the forefront of the profession. I remember Bernie Saffran's enthusiasm in recommending that book, back in my undergraduate days at Swarthmore.