The rich really are different
It’s understandable—this doesn’t make it correct or forgivable—that many of us unthinkingly assume that any given distribution resembles a bell curve. They’re so familiar and ubiquitous that even statisticians call them the “normal” distribution. Problems arise because trailing in their wake are a host of ready-made assumptions. This matters because the AmLaw 100 is not remotely a “normal” distribution; it’s a power curve, with a few big players, a lot more in the middle, and a long tail of smaller fry. This isn’t a technical quibble; it has teeth.
First and foremost is that a bell curve’s “average” is a highly descriptive number; it defines the central tendency of the universe under inspection. For power curves, no such thing holds true; averages aren’t just misleading, they can approach falsehood. Consider a few characteristics of this year’s AmLaw 100: (a) 10% of the group’s total revenue is accounted for by the top three firms; and another 10% by the smallest two dozen; (b) 25% comes from the top nine firms and 25% from the bottom 50; and (c) the top three’s combined revenue was over $8-billion and the bottom 20’s under $7.5-billion. In short, big firms really matter. Their performance can easily move gross measurements for the entire group.
Given this, it’s worth looking at the 100’s nominal increase in revenue ($3.5-billion) from a few other perspectives:
- How many of the 100 firms accounted for, say, two-thirds of that increase? The answer is about 20 firms; the other 80 didn’t grow enough in absolute dollars to make much difference, or else shrank outright.
- 18 firms reported their gross revenue decreased and 20 reported lower PPP; just as a matter of first-blush intuition, how does this square with what you assume when you hear +4.3% revenue growth and +3.0% PPP growth?
- And, to exemplify how a small firm can have eye-popping percentage growth without moving the gross number needle much, while a big firm lumbering along with the pack can have the same overall impact but no one notices, consider:
- Husch Blackwell was the #1 firm out of the entire 100 in revenue growth rate at +23.2% (they merged), adding $81-million to their year-before total.
- Skadden was actually below the “average” growth rate at 3.5%, but tacked on $85-million, or a rounding-error’s difference with Husch.
Finally, let’s try a drastic thought experiment: Suppose the smallest 20 firms disappeared in a flash; they would obviously be replaced by the largest 20 firms from the AmLaw Second Hundred. How would this radical set of “alternative facts” move the numbers? Not much. The smallest 20 AmLaw 100 firms generated $7,491,600,000 in revenue, and the largest 20 AmLaw Second Hundred firms $5,819,000,000, or a difference of $1,672,600,000. This sounds like a lot of money (it would be to me), but it’s actually only 1.9% of total AmLaw 100 revenue. Dropping one single firm (White & Case, #11, $1,631,000,000 in revenue) would have an equivalent impact.
Our Original Sin, if you will, in our annual congratulatory fest on the publication of yet another set of strong-looking numbers is to lump all the 100 firms together in some undifferentiated clump—as if they followed a “normal” distribution. Understandable, yes; and unforgivable. The AmLaw 100 is an eclectic menagerie.
And did I mention that big firms really matter?
Survivorship bias and its kin
Let me close with something you don’t see in the AmLaw 100—or any current performance financial ranking: Firms that didn’t keep up. The AmLaw 100, the Fortune 500, the S&P 500, are all rosters of the winners—winning as defined by the guardians of the pertinent index. “Survivorship bias” is the handy term for this, and it stands for “the tendency for failed companies to be excluded from performance studies because they no longer exist. It often causes the results of studies to skew higher because only companies which were successful enough to survive until the end of the period are included.”[3]
In 2012 on the occasion of the 25th birthday of the AmLaw 100, The American Lawyer conveniently reported that only 69 of the original 100 firms from 1987 remained in the ranking, implying a nearly one-in-three attrition rate. Wouldn’t it be instructive to see the “alternative fact” of the 2017 performance statistics for the 1987 AmLaw 100 firms? That would be cause for even inquisitive analysts to celebrate.
[1] A little over a year ago I invited readers of Adam Smith, Esq. to respond to a poll asking how fully their firm responded to The American Lawyer for purposes of the AmLaw rankings and a few related questions.. Albeit anonymous by necessity, a few data points from those who answered:
- Fewer than half the firms provided “comprehensive” responses; the majority responded “mostly,” “partially,” or not at all.
- Among firms not participating at all, the last time they did respond was more than three years ago (92%) or “never” (50%).
- Asked point-blank about their “confidence in the numbers’ accuracy,” only 23% felt “the vast majority of firms report fairly and honestly,” fully half believed it has “become a marketing and not a financial exercise,” and 28% “found it difficult to give credence to much of what I see in the numbers.”
[2] https://data.bls.gov/cgi-bin/cpicalc.pl?cost1=1000&year1=201512&year2=201612
[3] “CAGR for Dummies,” Adam Smith, Esq. July 2013: https://adamsmithesq.com/2013/07/cagr-for-dummies/ (emphasis supplied)
