Fights over advance notice bylaws are becoming more common; I previously posted about Paragon Technologies v. Cryan, and right after that, VC Will decided Kellner v. AIM Immunotech.  In both situations, boards were found to have been overly aggressive in drafting and enforcing their bylaws, and VC Will went case by case to determine which actions were permissible and which were not.  And since the dissidents had not fully complied with even the permissible bylaws, VC Will would not order that their nominees be permitted to stand for election.

This strikes me as such a difficult problem.  On the one hand, there are good reasons for these bylaws, as both the Paragon and AIM disputes make clear.  In Paragon, the dissident really was playing games about providing information regarding its plans; in AIM, the contest was a continuation of one spearheaded a year earlier by a convicted felon who tried to conceal his involvement.  So yeah, boards have really legitimate interests in ensuring that shareholders have full information.

On the other hand, the blue pencil approach – where the noncompliant bylaws are severed and the legitimate ones remain standing – strikes me as having the same problem that’s been identified in the context of noncompetition agreements.  In the employment context, Chancery does not blue pencil, because:

To blue-pencil the provision creates a no-lose situation for employers, because the business can draft the covenant as broadly as possible, confident that the scope of the restriction will chill some individuals from departing. If someone does challenge the provision, then the worst case is that the court will blue-pencil its scope so that it is acceptable. It also enables employers to extract benefits at the expense of employees by including unenforceable restrictions in their agreements. The logical result of such a system is sprawling restrictive covenants.  Accordingly, “[w]hile, in some circumstances, a court may use its discretion to blue pencil an overly broad non[1]compete to make its restrictions more reasonable, this court has also exercised its discretion in equity not to allow an employer to back away from an overly broad covenant by proposing to enforce it to a lesser extent than written.”

Obviously, the employment context is not the same as an advance notice bylaw – dissidents are not nearly as vulnerable as employees – but there is, I think, a similar set of concerns.  If Delaware courts blue pencil the bylaws back to what – ex post – is determined to be reasonable, then boards have no incentive not to draft the most lengthy and unreasonable bylaws they can.  The dissident then has to guess at which ones are and are not permissible, and comply with the right set of bylaws – reveal too much and make themselves vulnerable, too little and disqualify themselves – while mounting an expensive court challenge that won’t necessarily create useful precedent for the next set of creative bylaws.

One aspect of the AIM case highlights the problem.  In 2016, the board had adopted a bylaw requiring disclosure of “all arrangements or understandings between such stockholder and each proposed nominee and any other person or persons (including their names) pursuant to which the nomination(s) are to be made.”  In 2023, after having dealt with the convicted felon’s earlier contest, the board amended the bylaw to broaden that provision and capture associates and family members in a manner that VC Will found was unreasonable.  When the dissidents challenged the amended bylaw, rather than strike it entirely, Will decided to restore it to the narrower, 2016 version, and measured the dissidents’ compliance against that bylaw – the one that did not exist anymore.  The dissidents had not complied with the bylaw that did not exist, and that was one of the reasons why VC Will affirmed the board’s decision not to permit them to advance their nominees. 

Now, given the peculiar facts of this case, I’m not troubled by that result, but notice the bind this would have placed on a good-faith dissident slate.  They would have found it nearly impossible to guess what information was actually required of them.

I wonder, then, if the solution is to set some kind of blanket rule that certain types of bylaws are almost always going to be permissible.  If the board adopts bylaws that go beyond that floor, it does so at its own risk – if any additions are deemed unreasonable, all are struck, but the floor (to the extent adopted by the board) remains intact.  That way, dissidents know what the floor is for compliance, and there is no risk that, say, a felon gets through without any disclosure just because the board was overzealous.  Meanwhile, if the board has genuine need for information beyond the floor, it has an incentive to be judicious in crafting its additional requirements.