Glenn Davis is Director of Research at the Council of Institutional Investors. This post is based on a CII publication by Mr. Davis.
This post summarizes and paraphrases comments shared at a roundtable on executive compensation that CII organized on Jan. 9, 2018, in Denver, CO. Participants included representatives of asset owners, asset managers, employee unions, corporations and think tanks. By design, investor representatives constituted a significant majority of roundtable participants, while executive compensation consultants and proxy advisors were not represented. A full participant list is provided at the bottom of the post.
The roundtable was conducted under rules that encourage participants to speak freely, with no attribution to them or their affiliated organizations. Views conveyed in this report do not necessarily represent the views of employers of roundtable participants, CII, its board members or staff. CII Research Director Glenn Davis organized the roundtable, and edited remarks for use in this report
Magnitude
For several roundtable participants, the amount of compensation paid to executives (magnitude or quantum) is a broad concern. For others, concern is limited to outliers.
We will approach companies with the “you make too much” issue in cases where pay is so far off the spectrum that it’s just indefensible. Anything closer to the margin is a difficult conversation. But if we have some evidence that pay at the top is negatively affecting employee morale, or if top executives consume an absurd amount of operating income, we’ll raise it.
I like the idea of investors trying to come up with a line for quantum because companies are drawing that line, and it just keeps going up. It’s useful to start laying down boundaries.
We have to remember there’s a market for talent. Around the edges, we can have the quantum discussion. But our collective interests are in dealing with long-term incentives and dialing down the complexity we’ve all created.
Having the right CEO is critical, and boards are understandably averse to losing a good CEO. But is the pool really that small? Is mobility such that excessive pay is required to keep them? While there are some exceptions, the superstar CEO is a notion I think we are evolving away from.
Financial compensation is an extrinsic motivator. At some level, additional money won’t motivate an executive the way we perceive, and it might actually diminish motivation. We need to shift the discussion to what intrinsically motivates executives.
Pay is badly broken, and it’s not just design—it’s magnitude. Quantum is a problem even at high-performing companies. But I don’t know how we get the genie back in the bottle. How would you go about introducing some restraint? The whole system is set up to do what your peers do.
Yes, levels are too high. Why? Part of the challenge is that the total number is an open question. Do you look at what has vested? Do you look at realized pay? Do you make assumptions and look at estimated future payouts? There’s no simple answer, and investors are taking different approaches.
The amount of pay for beta is what drives the quantum issue, and it’s what drives investors crazy more than the fact that the numbers are large. Unless we get into the weeds and deal with design and complexity, we’re going to lose the race. And that’s what’s been happening for the last couple of decades.
Participants shared a range of views about whether and how they incorporate magnitude into voting decisions.
A lot of institutional investors deal with quantum through non-systematic, internal processes; you wouldn’t know quantum matters to them by looking at their proxy voting guidelines. If quantum is a concern, perhaps we could send that signal to the market by raising it in a more visible way, whether individually or collectively.
I think we need to do a better job of challenging compensation committee members who approve horrible products. It’s been our experience that the committee members are deferring to compensation consultants on the details.
We don’t have a bright line voting policy on magnitude, but we do use it as a triage for where to look. Over the last couple of years we’ve built out a methodology that creates a scale based on the size of the company and industry peers. We start with quantum as a place to find the outliers.
We screen for quantum on a percentile of total pay, adjusted for company size. But we don’t see a problem with paying on the upper extreme end if performance is on the upper extreme end.
I agree with setting bars for quantum as a screen for evaluating whether the board and the compensation committee are doing their job, which is to attract the best CEO they can, incent that person to achieve certain objectives and align that person’s interests with shareholders.
We currently apply a quantum test to short-term bonuses, but not to long-term payouts.
Our say-on-pay guidelines include consideration of the pay ratio. How we’re going to apply it? That’s not fully resolved.
Design and Complexity
Participants voiced a need to simplify compensation practices, and shared growing skepticism toward the premise that performance-vesting pay is inherently superior to other methods of pay.
How do you know when an incentive system is too complex? If your compensation committee members can’t explain to investors how it works, there’s a problem.
My issue is with this whole idea that every year executives basically get a do-over. Every year they get a new set of options or equity. Most of the time they get payouts ‘at target’ or above. Simplifying the way executives are paid is going to be a big part of ending these scenarios of shareholders losing and executives winning.
When you over-engineer, people are prone to gaming. The problem I see with pay for performance is that it works too well. You get exactly what you are incenting… Does this method work? It’s hard to pivot away from.
The current system isn’t as performance-dependent as we would like to think. Generally, unless there is a shock, you see rising pay year over year. And when the stock is down, executives get more shares. Companies are more focused on creating incentives looking forward, and so pay doesn’t reflect past performance.
Retention bonuses and awards for closing deals: If an M&A deal is great for shareholders, we’re in a rising market, and most of the executives’ holdings are in-the-money, why do we have to give them an extra incentive for doing their job?
Participants expressed an openness to focusing on base pay and time-vesting restricted stock or restricted stock units.
We should be talking about big ideas, not tweaking at the margins. A larger cash component and a long-term grant of time-based restricted stock is something we can all understand and compare. We don’t have to guess its worth. Companies perceive that investors won’t support a simple package of cash and restricted stock, but that approach has advantages that investors and proxy advisors may be forgetting.
With the changes to the tax code, we may see more companies in essence folding their annual incentive programs into salary—which in a way makes sense, in that annual incentive plans were never really that performance-based to begin with.
I believe fixed pay can work well for shareholders if you have a strong board willing to step in when the CEO isn’t executing on the long-term strategy. We need to solve the obfuscation issue. The problem we face is that shareholders are playing “whack-a-mole” with metrics. If you challenge them on one measure of performance, they have another. We are mired in debates about TSR vs. ROIC, absolute vs. relative, GAAP vs. non-GAAP. A return to more fixed pay would alleviate some of that.
We would like to see less emphasis on from performance-vesting shares and PSUs because they have inherent problems. Performance-vesting shares are difficult to design properly and they don’t capture all the drivers of business models. There isn’t much empirical evidence that performance-based awards have increased alignment of pay and performance. Also, performance shares are typically disclosed at target rather than max; an RSU will be disclosed at grant date value, making it easier to extrapolate what the quantum is.
Executives should be compensated with a combination of cash and restricted share units vesting over the long-term of five to 10 years, ideally closer to 10, with no accelerated vesting. Compensation committees would have discretion to balance cash and RSUs. Performance-based targets could color decisions about how many RSUs to award the CEO in a given year.
There are mixed views on how replacing performance-vesting shares with restricted stock units would impact quantum. CEOs may demand a higher number of shares for long-term vesting schedules. On the other hand, if there is certainty that shares would be distributed at a given point in time, it may decrease quantum.
Ideally, upside and downside should be felt equally. With an RSU, you’ll see less risk appetite in certain companies—certainly less than with stock options. But that is a consideration that needs to be thought through.
I think PSUs have been overdone and a lot of investors are becoming more skeptical because of what they’re hearing from companies. It just hasn’t filtered to proxy advisors yet.
Participants expressed support for boards’ discretion to select various award types and multiple time horizons.
Annual awards motivate day-to-day behavior, and are an important part of long-term shareholder value. We need a balance between short-term and long-term incentives, but the longer-term piece should be prominent—particularly in cyclical industries. But we also need to remember that the terms of long-term incentive awards are at risk of losing relevance over time, and recipients apply a discount to them due to their uncertainty.
All companies are different, and yet we’ve seen homogeneity in the way they’re structuring their packages, either because of investors like ourselves or proxy advisors’ guidelines. I would like to see programs designed individually by companies.
Right now many companies offer multiple types of awards without thinking through what makes sense for their business. Companies should step back and ask, “What’s the right vehicle for our company based on the industry we’re in, the length of our cycle, our peer groups?”
Participants’ views were mixed, but generally cautious, on hold-to-retirement requirements on vested equity.
I support holding a significant portion through retirement, 50% or 75%. Without it, the alignment goes “poof.”
There’s a cottage industry of diversifying through derivatives, and I have a libertarian streak that doesn’t like binding someone from selling their property once it’s vested.
Policies of very long vesting—six, seven, eight years—get the benefit of holding requirements without the problems of restriction.
Participants said the fragmentation of investor views on executive compensation is a challenge, and that a more collective voice from the investor community, if possible, would facilitate evolution in executive pay design.
A lot of investors, proxy advisors and companies are open to change, but sometimes change doesn’t happen until there is a collective view.
If you talk to 30 different investors, you may get 25 suggestions on pay design. Performance-based long-term incentives are important to some, so it gets reflected in the design.
I support less differentiation and more collective action by investors.
In some ways we’re shooting ourselves in the foot by allowing all this complexity. It results in owners with fragmented positions, making it more difficult to approach companies as united. All of this feeds on itself. Coming up with some structure is helpful.
We don’t expect change overnight, but if more investors can be explicit with their views, more companies can have confidence that they can make changes without facing a backlash.
I hear different critiques of pay, and I agree with all of them. Right now we support performance-based pay tied to the business plan. We all have concern about structure but I am sober about whether the investor community can speak as one.
Transparency
Participants generally expressed a desire for specific information that would help shareholders assess the rigor of executives’ incentives.
What we really need from disclosure is the ability to judge rigor on a going-forward basis. When the metrics are available us, we can discern whether stretch goals are valid. Voting is hard because it’s difficult to judge with the information we have whether the plan is likely to work or not. I am not motivated by the competitive harm argument for keeping this information private.
There are many investors who are voting based solely on the output. I am not comfortable voting on a plan when I don’t know what’s in it. Some companies opt not to disclose targets, metrics, thresholds or the upper limits of payouts. Some companies disclose but provide no view into what has changed. They’re immunizing themselves from judgment.
Should we push companies to disclose? Should we press advisors to penalize companies for not disclosing? If we had a united front on this, perhaps we could get more valuable information. Quality decisions are difficult with little time and little information. I think we need to be asking the proxy advisors to help us.
If you’re encouraging companies to be creative and company-specific about metrics, there should be some accommodation that disclosure is not mandatory.
Every industry has two or three metrics that most analysts pay attention to, that companies are focused on. Yet we don’t always get that information from companies, even after the fact. I think that’s the big piece.
Advisory services tend to hone on the idea of “Is the performance roughly analogous to peers, and is compensation roughly analogous to peers?” What’s missing is an analysis of how “stretch” the goals really are. If the goal for the annual bonus plan is being set at 40 percent of what the performance has been year after year, that’s not a bar. That’s a joke.
Participants shared nuanced views about the objective of the Compensation Discussion & Analysis (CD&A) section.
The CD&A should be used to help investors understand whether the board and compensation committee are doing their job.
The CD&A is expanding, but I’m not convinced it’s any easier to understand how plans actually work.
The CD&A is an opportunity to get away from detailed aspects of pay and communicate corporate culture and broader compensation design.
Purpose and Meaning of Say on Pay
A discussion on the advisory vote generated vigorous debate on how effectively the shareholder community is using the tool, as well as questions on how effectively some companies interpret voting results.
Shareholders now have three powerful opportunities for reform aside from exiting their position. They can vote against compensation plans, they can vote against directors and they can vote against on the advisory vote. Shareholders need to start looking at themselves and ask, “What are we supporting, and what are we incentivizing?” We can’t complain about compensation and not do what’s in our power. I don’t want us to absolve ourselves of our active role in this, and I think that means voting more discerningly.
Whenever we vote against on say on pay, we vote against compensation committee members as well. But I don’t necessarily see that from everyone else. When directors see their names next to unusually high “against” votes, they take that more to heart than an advisory vote, which is less directly attributable to them.
There seems to be some consensus that there’s a problem with pay broadly, including with boards acting in good faith—that there’s something out-of-whack. If that’s your view, the tool you have is your vote. Why are investors reticent to use it?
Companies and boards are wise to focus more attention to say-on-pay results on a percentile basis relative to their peers. A lot of board decisions about executive pay are justified by, “Everybody’s getting this, and we’re getting high votes.” But the raw percentage of support doesn’t tell the full story.
There is a fundamental disagreement about the purpose of say-on-pay. In my view we’re doing more good by stamping out the bottom 10 or 20 percent than by impractically voting against when we don’t see perfection.
Nowhere near 10 percent of companies lose their say on pay vote. If we actually reached the bottom 10 or 20 percent, that would make a difference.
We were initially skeptical of say on pay because it is such a blunt instrument. But we’ve come around to see that it’s a source of influence and engagement with the company. Our votes against have dropped, and we maintain access and relationships with management. We may have issues with compensation. But if we’re talking to management and we feel they’re responsive, we don’t see the need to follow that up with a negative vote.
The responsibility to render a decision on all say-on-pay votes is complicated. It forces you to analyze companies where time could be better spent elsewhere. The advantage of a red flag approach is that you can spend more time where it’s likely most needed.
The social dynamic is just different if you own 5% as opposed to 0.5%. The worry is that large holders are inclined to be supportive to keep companies off their backs. Pressure and communication from the company can be intense. But that communication can also lead to understanding companies’ situations in ways broadly diversified investors don’t.
Participants conveyed that the first six years of the mandatory advisory vote led to some unexpected outcomes.
Say on pay hasn’t failed us. In many cases we’ve stamped out egregious practices and the low-hanging fruit that was causing misalignment. Engagement has meaningfully improved. The biggest impact of say on pay wasn’t pay design or quantum, but on prompting more engagement.
To some degree, say on pay has insulated compensation committees by diffusing pushback from shareholders.
It shocks me how much engagement increased following the arrival of say on pay. Were companies comfortable when compensation committee members bore all of the criticism?
Say on pay is an early warning sign—a release valve for pressure in terms of when boards need to make changes. Whether the vote passed or failed is overemphasized.
When our beneficiaries see high compensation, it’s hard to explain to them why so many shareholders were supportive. In our minds, we feel like say on pay hasn’t gotten to quantum, and it hasn’t gotten to long-term alignment. But it’s a tool for dialogue.
Participants indicated that even when a company outperforms, their unwillingness to support faulty executive compensation design doesn’t waver.
If the process or structure is flawed, we vote against even when there is alignment. The outcome may look like alignment this year, but we have to be thinking about the future.
Compensation comes straight out of shareholder money. If you’re seeing millions going out the door, and the performance isn’t there, it’s easy to explain to our plan participants. When there are aspects of pay programs that are destructive, we have to vote against even when performance is there. In those cases, it’s especially important for us to explain our basis to our beneficiaries.
* * *
Participants
Cambria Allen, UAW Retiree Medical Benefits Trust
Donna Anderson, T. Rowe Price
Ken Bertsch, Council of Institutional Investors
Ben Colton, Norges Bank Investment Management
Jennifer Coulson, British Columbia Investment Management
Brian Denney, Vanguard Group
Tom Elliott, Capital Group
Michael Garland, New York City Pension Funds
Trevor Gunderson, General Mills
James Hamilton, BlackRock
Matthew Jacobs, California Public Employees Retirement System
Dale Johnson, Los Angeles County Employees Retirement Association
John Keenan, AFSCME Employees Pension Plan
Matthew Leatherman, FCLT Group (Focusing Capital on the Long Term)
Renaye Manley, Service Employees International Union
Aeisha Mastagni, California State Teachers’ Retirement System
Hope Mehlman, Regions Financial
Bridget Murphy, Washington State Investment Board
Jennifer O’Dell, LIUNA Staff and Affiliates Pension Fund
Miguel Padro, Aspen Institute
Fionna Ross, Aberdeen Standard Investments
TerriJo Saarela, State of Wisconsin Investment Board
Paul Schneider, Ontario Teachers’ Pension Plan Board
Margriet Stavast, PGGM
Tracy Stewart, Florida State Board of Administration
Ted White, Legion Partners
Also in attendance
Luz Rodriguez, Colorado PERA
Amy Borrus, CII
Marie Brodmerkel, CII
Glenn Davis, CII
Jeff Mahoney, CII