Most compensation committees review information gathered from a customized peer group of companies as part of compensation program development and relative performance reviews, and peer group construction is often a matter of significant debate.

Common sense would suggest that if the goal is to understand what is competitive pay to attract and retain and motivate executive talent, then the peer group roster should include companies of generally similar industry, ownership status (e.g., public vs. private), and size or complexity. Narrowing the roster to isolate those companies most similar with respect to business model and performance is ideal, but must be balanced with the need to identify a sufficient sample size to minimize the impact of data aberrations.

However, what benchmark data is used for also has implications for how the peer group is constructed.  Consider the case of companies considering an IPO in the next 6 to 18 months. Critical compensation activities for these companies often include:

  • Developing an equity plan with an appropriate share reserve to cover intended future grants that is also compliant with regulatory requirements and aligned with shareholder expectations;
  • Designing special equity awards for executives that signal to the investment community that management is appropriately aligned with their performance expectations and shareholder value creation—and has significant skin in the game to foster retention post-IPO;
  • Implementing severance and change-in-control benefits that help to avoid business disruption or distraction in the event of deals that—while creating value for shareholders—may place employment at risk; and
  • Recruiting outstanding outside directors with the attendant additional requirements and risks of public company governance.

The most useful information for supporting these decisions will be gathered from companies who have confronted similar challenges recently. In other words, the experience of other recent IPO’s is more relevant than the pay programs of companies that have been publicly traded for 5 years or more. This benchmark data is more likely to be sourced from S-1 filings than from Proxy Statements, and filtering mechanisms may include factors such as:

  • Industry, but also considering “related” sectors (think two-digit GICS classification rather than four-digit)
  • Recency of IPO (e.g., past two to three years)
  • Market value at time of IPO
  • Size/complexity at time of IPO
  • Growth trajectory leading up to IPO
  • Founder status—and overall ownership position—of management prior to IPO

Those familiar with the voting recommendation reports from ISS and Glass Lewis will note that this approach to peer group construction is quite different from the protocols used by these two proxy advisors. While each firm has periodically modified their approach, and Glass Lewis will be announcing their latest updates shortly, they have generally focused on companies of similar size (revenues are the primary lens), industry (GICS classification is a key consideration) and “peers of peers” analytics.

Keep in mind that the majority of IPO’s are eligible for “emerging growth company” (“EGC”) status[1], and therefore may avail themselves of scaled back executive compensation disclosure requirements. They also are not subject to Say-on-Pay votes until they emerge from EGC status. Practically speaking, this means there is no requirement—and in fact, it is rare—for an EGC to disclose to shareholders in either its S-1 or subsequent Proxy Statements the roster of companies used for executive compensation benchmarking purposes.

We have found that post-IPO, companies tend to shift towards a more traditional approach to peer group construction. This is an organic evolution as committees face new challenges, such as considering what are reasonable annual pay opportunities and how to test the alignment of pay with performance. By the time a company is subject to Say-on-Pay, it is unusual not to have substantial overlap in peer group rosters as compared to those compiled by ISS and Glass Lewis (excepting certain special situations such as pre-commercial biotechs and / or non-US public companies).

In any event, the first question in peer group construction should always be “what is the business decision that this data is intended to help us solve?” Different rosters may be—and, in fact, ought to be—constructed for different decisions.


[1] An issuer is eligible for EGC status until the earliest of (1) the last day of fiscal year during which it has total annual revenues of $1.07 billion or more, (2) the last day of fiscal year following the 5th anniversary of its IPO, (3) the date on which it has, over the previous 3-year period, issued more than $1 billion in convertible debt, or (4) the date it becomes a “large accelerated filer” (generally when its public float reaches $700 million and it has been filing Exchange Act reports for at least a year).