The Office of the General Counsel for the National Labor Relations Board spends much of its time handling charges of unfair labor practices (ULPs). An excellent report in 1999 by the General Counsel, Fred Feinstein, helps us think about quantifying legal complexity. [www.lawmemo.com/nlrb/gchighlights.htm]
ULP determinations had become more complex. The report explains: “This is demonstrated by the fact that, since 1991, the average length of the transcript of a ULP hearing has increased 30 percent, from 473 pages to 616 pages.”
What sparks fly from striking that average? (1) We need numbers collected consistently over time to have any chance at putting a numeric face to legal complexity. (2) We need to consider exogenous forces that might have driven the numbers, other than our conclusion. For example, did procedural rules for hearings change, did the ULP bar move to larger law firms, were court reporters compensated by the page instead of flat fee or were they changing page margins? (3) We need confidence that the average of pages more accurately represented hearing length than the median. Suppose each year one monster hearing drove up the average, but the quotidian transcript didn’t budge. (4) We need to accept that page length signals complexity (not mere prolixity).
After taking such queries into account, a law department might assay its own complexity indicator. It might look at changes over time in the length of EEOC charges, leases, confidentiality agreements, or patents. Looking back, when the numbers were what they were, unconsciously lying there, so to speak, the department can use them with defensible confidence.
Once the department uses those numbers going forward, however, it faces the corruption of gaming. Members of the department will change their behavior to look good under the new measurement system. If “complexity” is honored, as an example, and if it is measured by pages, better plant a forest!