After all these years, I find it amazing. Consideration of executive rewards is still split up into pieces. And, those pieces are handled by different internal functions and by different consulting constituencies.
In a fairly typical case, cash, long-term incentives and equity are handled by the executive compensation function and by the executive compensation consultants. Executive retirement programs are typically handled by the retirement function and by the retirement consultants (frequently actuaries).
This is not a problem. The problem lies in the fact that the left hand and the right hand don't communicate with each other. And, they don't have compatible methodologies.
Let's look at retirement first. Traditionally, executive retirement packages have been designed to replace some targeted percentage of the executive's base plus bonus in their last few years before retirement. That methodology is not wrong. In the typical executive retirement study, consultants are asked to benchmark the plan design. Does it align with current trends and practices?
Consider executive compensation. Here, consultants look at such this as total cash compensation and total direct compensation. They benchmark this against the organization's peer group regressing (adjusting) for differences in size (and sometimes complexity). They develop medians and percentiles. That methodology is not wrong.
Suppose a Board chooses to pay its CEO at the 60th percentile. Perhaps they feel that their is complexity to their organization that belies its size. Suppose they also have an executive retirement program that their consultants say is pretty mainstream. I am going to tell you that almost to a degree of certainty, the retirement consultants have not considered the level of the CEO's pay in determining that the retirement program is mainstream. Isn't deferred compensation a part of compensation?
What would happen if we used the same approach for retirement benefits as we do for other forms of executive compensation? Suppose we calculate an annual value for such benefits and add it to other forms of compensation before doing that regression. Something tells me that the results might be surprising. In some cases, it might justify that rich SERP for which the proxy analysts have such disdain. In other cases, we might find that the company is perhaps inappropriately inflating TOTAL compensation -- the sum of the value of the entire rewards package.
In order to make this work, the executive compensation people need to talk to the retirement people and conversely. They need to speak each other's languages. Today, there are many gaps. There just aren't enough of us who are bilingual in this regard.
Perhaps we need to be.