1) The firm uses a two-dimensional grid to determine bonuses. The rows of the grid are billable hours ranges (in increments of sometimes 50 but no more than 100 hours). The columns of the grid are class years.

2) The firm benchmarks the grid to the New York market by setting 1800-1900 as what is needed to get the New York market salary (35, 40, 45, 50, 55, 60). Therefore an associate gets at least the NY market by billing 1800-1900.

3) The firm does not release the grid and only equity partners know all the data points on the grid. At a meeting the firm will reveal generalized information about the grid like the high's and low's for each class and perhaps the number of people in each class X who get in the 130,000's, 120,000's, etc.

4) Associates are told their individual numbers in private. Nobody except the equity partners know the full.

5) Posts like http://www.infirmation.com/bboard/clubs-fetch-msg.tcl?topic=Greedy%20NY&msg_id=0035YR are attempts by people to get various data points from their colleagues and piece the grid together. They are accurate in some respects but are too high in certain instances and too low in other instances.

6) The low sub-market numbers you are seeing are for laterals who were not with the firm for the entire year.

7) The way the meritocracy works is to allow an associate to be "bumped up" 1 or 2 class years on the grid if he/she receives an "above the class" or a "top of the class" rating, respectively. (An example of how the meritocracy works: a 3rd year who is above the class and bills 2200 hours will get the same bonus as a 4th year who is with the class who bills 2200 hours.) Essentially all of the junior associates are rated with the class. Associates begin to break out above the class in the mid-level years.

8) It seems this year that an average billing junior associate (the firm's average is about 2050) gets about $10-15K over market and an average billing mid-to-senior associate gets at least $20-30K over market. As I explained in 5, the precise accuracy of numbers is hard to verify as it is based on a collection of data points and not the full grid.

9) Of course, you can shatter the NY market and make a ton of money on the grid by being either a high biller or being rated above the class.

(http://www.autoadmit.com/thread.php?thread_id=549037&forum_id=2#7256881)