DealLawyers.com Blog

Over on the Harvard Governance Blog, S&C’s Melissa Sawyer has a recent post addressing something that’s become apparent to anyone who has been around M&A for a while – merger agreements have become much longer than they used to be.  Melissa attributes this phenomenon to the common practice of developing a “reference set” of the other party’s (and their counsel’s) precedent agreements and the tech tools – now supercharged by AI – that allow deal lawyers to “cobble together Frankenstein drafts from different sources within the reference set in record time.”

She says that this practice has had two detrimental effects on the quality of drafts – “feature creep” and “recency bias.”  This excerpt explains:

Feature creep.  “Addition by subtraction” is virtually unheard of in M&A drafting.  Document assembly almost always results in supplementation: more “belts and suspenders”, more “provided howevers”, more “for the avoidance of doubts” and more “notwithstandings”.  For example, the 2009 ABA Public Company Target Deal Points Study analyzed five potential carve-outs to “material adverse effect” (“MAE”) clauses that had become customary at that time.  That list of customary carve-outs had more than tripled by the time of the 2024 study, with many MAE definitions now covering over a page and a half of dense text.  With scant evidence of reliance on these carve-outs in the real world, the utility of all this extra wording is largely untested.

Recency bias.  Examples of recency bias abound in merger agreement drafting.  Since COVID-19, almost every merger agreement has included explicit pandemic-related clauses.  After the publication of an influential article in 2014, almost all merger agreements have defined the term “Fraud”.  Since the height of the #metoo movement, many merger agreements now include sexual misconduct reps.  The triggers for new provisions are variable – Delaware judicial opinions, geo-political events, broken deals, academic literature – but the results are the same:  more words on the page and more time spent negotiating whatever issue is the latest flavor of the day, often without regard to whether those issues are likely to be material to the parties and in some cases long after the waning of the socio-political discourse in which concept originated.

Melissa posted a link to this blog on LinkedIn, and the responses to her post are worth reading.  For example, John Coates added a link to his own analysis of the growing length of merger agreements, while Joung Hwang asked whether clients would be willing to pay for the effort it would take to trim down the typical merger agreement.

My favorite response comes from Enam Hoque. His decision to quote (“yes I said yes I will Yes”) from Molly Bloom’s 4,391-word, one sentence soliloquy that ends James Joyce’s Ulysses in response to Melissa’s question “Do you agree with me on this?” works on many levels and wins the Internet for today.

– John Jenkins