The Advisors' Blog

We’ve posted the transcript for our recent webcast “Equity Award Approvals: From Governance to Disclosure” with Jeff Joyce of Pay Governance and Sheri Adler and David Kaplan of Troutman Pepper Locke. They addressed:

1. Not Your Kindergartener’s Math: Share Counting
2. Planning Ahead: Award Design
3. Approval Formalities:
   • Who Approves?
   • What Gets Approved?
   • Grant Timing, Sizing and Disclosure
4. Documenting and Communicating Awards

Sometimes I find myself afraid to listen to these sorts of foundational programs because I know I’ll learn something that I will feel like I should already know. That was definitely true for me in this program, so here’s an excerpt of one of the discussions I found really illuminating. I hadn’t previously focused enough on the size of the impact of this conversion method — and the potential consequences for pay-for-performance assessments (or how explicit good disclosure is on this point).

David Kaplan: [I]f you’re granting PRSUs, you’ve got a choice between: Am I going to convert that target dollar value to a number of shares based on a stock price (spot price or a multi-day average) or using a Monte Carlo value? . . . It’s not unusual in my experience to see Monte Carlo values that are 130%, 135%, 140% of stock value. Jeff, are you seeing values far beyond that, or is that what you see typically?

Jeff Joyce: I think that’s the typical range, but I would note that there are some companies that do have Monte Carlos in the upwards of 150% to 160% of the face value of the stock, and that’s where this typically comes into question.

When you look at prevalence of practice, and this was somewhat surprising to me, a good portion of the market that uses PRSUs based on a market condition, a stock price or a TSR condition, uses stock price on the grant date rather than Monte Carlo for determining the number of shares . . . Basically, they say, “That’s too confusing for our participants. We’re going to use the stock price on the data grant just like RSUs,” but as you point out, there’s implications to doing that.

The most notable is for your named executive officers, you’re granting a target value using a lower share price so you get a greater number of shares, but then when it’s disclosed in the Summary Compensation Table, you apply the Monte Carlo, the accounting value, to that and it’s a value greater than that $1 million. It would be 1.5 times that if the Monte Carlo is 1.5. There’s a disconnect between the intended value and the disclosed value and the accounting expense associated with the award. If you’re using market data that doesn’t take into account the valuation approach, you’re over-delivering from a competitive perspective . . . I think a key takeaway from today’s discussion is to understand how shares are being determined for grant and just ensuring that the committee is aware of that, because from a Say-on-Pay perspective, this does influence the proxy advisor pay-for-performance assessments.

David Kaplan: . . . As you pointed out, Jeff, these differences can be material. If you’ve got committees that are saying, “We want to pay our named executive officers at the 50th percentile, and we’re going to convert PRSUs based on stock price,” and you’ve got that 150% Monte Carlo in your example — if your program is heavily tilted toward performance-based comp, as we’d all advise it should be, when you end up applying those Monte Carlo values, the accounting values come way higher.

You can end up essentially having total pay to CEOs be, I don’t know, 20% higher than really benchmarked. What you thought was a 50th percentile pay package turns into a 75th percentile pay package or more. The message to the boards we advise is: Use whatever conversion value you’d like, but know that you’re doing that. Understand those outcomes and act accordingly.

Members of this site can access the transcript of this program for free. If you are not a member of CompensationStandards.com, email info@ccrcorp.com to sign up today and get access to the full transcript.

– Liz Dunshee