Matt refers to an old paper he wrote in 2013 - check it out here
The Larcker/Tayan ECGI paper is here
TRANSCRIPT
Matt Voiceover
Welcome back to Sound-Up Governance. Today we have another episode that's just me. But wait! Before you shut off your computer or phone or whatever, I want to talk through a super important topic that I honestly believe is completely broken: executive compensation. Especially CEO compensation. I might be wrong, but if I am, then I'm wrong in a pretty well informed way. In other words, you might find yourself disagreeing with me and learning something. I'll give away the headlines now. First, I don't think anyone anywhere has any idea what good CEO compensation looks like. Second, as a result of the first point, all the complexity we've built around CEO pay is basically lipstick on a actually, I've never really liked the lipstick on a pig expression. Pigs are already very adorable. What's an objectively ugly critter? I honestly don't know. But my handsome tarantula, Lumpy, would look pretty weird with lipstick on so let's go with him. Third, I think part of this whole CEO pay nonsense could be solved if we completely redesigned the role of executive compensation consultants. Anyway, let's start at the beginning.
Matt
I'm actually kind of excited to talk about this. So believe it or not, there was a time not that long ago where there weren't a lot of people or a lot of firms or organizations that had good data on executive compensation. And way back in the early chunk of my career at the University of Toronto, I ran a research centre called the Clarkson Centre for Board Effectiveness. And we kept on we spent a lot of our time looking through public filings, proxy statements, or what we call in in Canada Management Information Circulars and looking at executive compensation and looking at all that sort of hand wringing in the news, especially around the alignment or misalignment between pay and performance. And we thought, Well, why don't we gather some data and see what we can find? Because there weren't a lot of organizations or any that we knew of at the time, especially in Canada that we're doing that. So we built this really amazing, really, very unusual data set, where we took all of the CEO pay data that we could find, and normalized it so that it was all comparable from company to company. So for instance, we valued the options ourselves so that we weren't reliant on the issuers' valuations, and we valued the sort of medium and long term compensation the best we could so that we weren't reliant on their valuations just so that we could have an apples to apples comparison year over year and company to company. So it was a really cool, very robust data set. I think, by the time we stopped gathering this data, we had something like 20 years of historical stuff. It was super cool. And so what did we do with it? We did a couple of interesting things. One was we examined this pay/performance length that everybody was so concerned about. What we tried to understand was over which time horizons, or over multiple time horizons, to what extent was CEO pay aligned with specifically return to shareholders. So what we looked at was percent change in CEO pay over time, versus increase or decrease in total shareholder return. And so we wrote a few papers we looked at at firm level stuff we looked at at index or system level stuff. Our Rotman colleagues, Alexander Dyke and Craig Doidge and Hamed Mahmudi and Aazam Virani wrote some really cool papers on it. And so you can look those up. They're kind of interesting about like shareholder activism and CEO pay and that kind of stuff. So ultimately, I mean, who cares about any of this except for wondering, what did we even learn? Well, what we learned and there's a I'll put a link to a paper that I wrote. This was literally 10 years ago, I think it was 2013. And really, what we found was pay and performance, for the most part, are aligned over time. Why? Because they both go up all the time. I mean, there of course, are going to be weird years where you have outliers in terms of CEO pay because of unusually big event driven grants or you have a really strange year in terms of performance. But when you look at it over three, five, ten, or more years, it all smooths out and CEO pay and return to shareholders just kind of always go up. It's not that interesting.
Matt Voiceover
So that's where things started for me on this topic, I should admit here that I've always been mostly ambivalent about the question of the right amount of pay a CEO should get. Don't get me wrong, I think income and wealth disparity are a significant driver of a lot of economic and social issues. When I say ambivalent, I mean I have no idea what the right amount is to pay a CEO. And I suspect we could be making a compelling argument for just about any amount we could put on the table. But I got a whole new context and vibe around this ambivalence in a single moment, some 10 odd years ago.
Matt
Back in the day, some of my colleagues at the Rotman School of Management, the folks who were working on corporate governance research and education and so on, would invite a group of professors from Stanford, because Stanford had, at the time, probably North America's most interesting group of researchers looking at corporate governance. And so they'd invite the Stanford guys to Toronto to hang out. And there would be this event where the Stanford guys would kind of run sessions and have panels of CEOs and corporate directors from Canada, and lead conversations that were kind of rooted in the research that they were doing. And I remember, I didn't know much about David Larcker. At the time, this probably would have been 10 years ago, I have become much more familiar with his work and really like him. He's just recently retired. But I'll talk a little bit more specifically about his work in a second. But I remember at one of these Stanford/Rotman days, he was doing a short presentation on some of his work. And he made a comment that I think he just intended as a bit of a throwaway. But it made a really big impression on me, he said, "you know, no one's ever studied how low CEO compensation can go before a CEO will quit. And so because we don't have that floor, the variable of the floor, we don't really know anything about what amounts are right." And I thought, this is a really important comment, because at the time, I was, you know, spending a lot of time gathering data and looking at trying to figure out what the right questions were to ask of the data. And we were really interested in pay performance alignment. But I started, it sort of planted a seed of skepticism in me about how I should be doing this type of work, or whether I should be doing it. And by this type of work, I mean, even examining CEO pay in the first place, because if we don't even know what good looks like, or have the opportunity to study it, then who cares? But anyway, fast forward to 2023. David Larcker has done this he put out as part of his wrapping up his amazing career, he put out a paper, the name of which I'm blanking on, give me a moment, I'm going to find out. [Intermission chimes] Okay, this, I found it, it's great. It's, it was published by the European Corporate Governance Institute, ECGI, May 2023. And it's David Larcker and Brian Tayan, sorry, Brian, I don't know how to pronounce your name. And the title is "Seven gaping holes in our knowledge of corporate governance," and they're all really good. And they're all they all should plant a seed of skepticism in anyone who read them. In fact, the first one is, "number one, effective boards," and they basically say that we don't know what characteristics make a board effective, despite all the research and dialogue and, and barrels and barrels of ink and millions of dollars that have been spent on education and research, and so on. So it's a really great point. But more relevant to what we're talking about, is there's there's a few different points here about executive compensation, but and one of them, he says, "despite the very large amounts paid to the CEOs of the largest US companies, we simply do not know the value of the CEO to an organization and what pay levels are appropriate for this employee." And I think this is a really, really cool and important point, because we talk a lot. And I mean, I know a lot of really, really smart people who do executive compensation consulting or sit on compensation committees. And we talk a lot about the amounts and benchmarking and making sure that it's well positioned in our peer group and making sure that we've got a good mix of short and medium and long term incentives and metrics and pay categories and making sure that the amounts are right, and shareholder alignment is robust, and that we've got internal and external goals and ESG and, and, and, and, and... But turns out, we don't actually know. And there's no research that supports any position on what amounts are actually right. And I'm gonna make an argument here, and I know that this is a bit half baked. ached and stupid, but hear me out. I just did a rough calculation before sitting down to record this and the average S&P 500 CEO is paid roughly 278 times the average US worker's pay per year. So if we can agree that short term short term share price hiccups aren't really the same as the value of an employee. So for example, like if a CEO just disappeared, yeah it might have a short term impact on the share price. So if we can agree that that isn't the same as the value of a CEO, I think it's hard to argue that a CEO adds more value to a firm than 278 other employees. Right? So for example, if a CEO did just magically disappear, the firm still goes. In fact, there's people like that it barely would matter. Not to say that CEOs don't matter. But everything would keep on happening. If 278 employees disappeared from, let's say, a factory, that factory stops working, right? So I think that there's this, there's an important thing here to realize that CEOs are super replaceable. They're almost as replaceable as anybody else, especially because there's lots of other people around who are interested in doing that job. I think. Although Larcker made another point in this, this amazing parting paper, that in a... sorry, I'm ignoring Brian Tayan here. Larcker made this great point about CEO market inefficiency, meaning CEOs... we have no idea how replaceable they are. We're making an assumption that the supply and demand for CEOs is roughly equal, like there's an efficient market. And we just don't know, right? We don't know the nature of the quality of the applicants or their potential impact, we don't know... we don't have any visibility into the access to internal and external candidates for any given firm. So we just really have no idea, it could be that every single CEO is easily replaceable. We do know that that - and Larcker makes this great point and it's been studied - we do know that firms are way too slow to get rid of underperforming CEOs. We do know that they're way too slow at getting on effective succession planning and so on. The system is really, really kind of messed up. So I guess my argument here is, we really don't know anything. So trying to pretend like we can optimize CEO pay through some combination of tax optimization and benchmarking and so on. It's just a fantasy. It's really completely made up. Sorry.
Matt Voiceover
Okay, so whether this is all resonating with you or not, I've recently had a thought about the executive compensation consulting industry. Let me just say that I'm probably wrong. And there are so many super smart executive compensation consultants out there, some of whom are among my favorite people in the governance space. And they're probably more right than me. But, and it's a big but, I think they might be missing their biggest chance to have an impact. Let me explain.
Matt
I think boards, first of all, should be wondering if they should be taking a completely different and new approach, right? Because the status quo is based on nothing, there's no evidence that we have it anywhere close to right, or that it's working the way that it should. So let's start by, and I want to reflect on this industry that's grown over the past 20 years especially around executive compensation consultants. So what in theory is an executive compensation consultant for? right? So I guess the high level thing is they're supposed to be helping to design incentives that help to ensure that the CEO's behavior is in line with the desired performance of the corporation and the interests of specifically shareholders, which is what where most of the alignment happens. And they do that by collecting data and generating benchmarks and crafting peer groups and perpetuating this philosophy of attract, retain and motivate, which we know as we've already discussed, is mostly just based on nothing. Right back to the sort of inefficient market or the assumption of the efficient market thing. And they're also the executive compensation... Executive compensation consultants are also there to help boards to make decisions about CEO pay. And those decisions are more complex, especially now than the boards can completely understand on their own. So it helps to have someone giving some advice on that. And the consultants are also there to help organizations, especially boards, to signal to investors that the boards have independent advice so that they can disclose in their proxies "Hey, we hired so and so consultant. And that means that we approached executive compensation in an independent and hopefully non conflicted way." So what's the result of all this? Is, as I found in my original work that I discussed before, pay and performance, just both go up and up and up and up and up and up forever. But we still don't know what the floor is, or the value that a CEO creates for a firm. So we've got this pay that just goes up and up and up. But we actually have no idea if that's the direction it should go. Plus perpetuation of wealth and income disparity, which as I've mentioned before, I kind of feel are big deals, both economically and socially. Big negative deals. So if I have all these problems with the sort of status quo of executive compensation consultants, and boards and the decisions they make together, I can't just complain about it. I should put an idea on the table. So here's my idea. I've got an idea of what executive compensation consultants could be instead. I heard actually just yesterday an example of a CEO, who just really wanted to spend more time with his son who had special needs. So what if the point of an executive compensation consultant is to find that? So that's just one example. Like, what if we find out... the job of the consultant is get to know the CEO so well and intimately that we know exactly what's going to eject them out of bed every day excited, and happy to do a great job and to empower others to do a great job. And also to understand what makes them go, "Ugh, I don't want to do this job." So that we can avoid that. And what if it is that they just want to spend more time with their family or with a particular set of personal situations, and we could find a way to make that dream come true. And to find the set of metrics and objectives over whatever appropriate timeframes that would be good for the organization and exciting for the CEO to achieve and aligned with the interests of shareholders and other stakeholders. To figure out what achievements we need to have to make that dream come true. Or any other dream. What if they want to play at Wembley Stadium with Slash in front of 100,000 people, or have a year to micro dose and explore their essence in the rainforest or whatever? What if that were the executive compensation consultant's job, right? And the other thing is that we've never really explored any counterfactuals. Like, we have so many 1000s of examples of firm years, where there's a single, super well paid CEO. And some of those firm years were amazing. Some of them were meh. Some of them were awful, right? So we don't really know if that model of a single super well paid CEO is good, or meh or awful. And we have no examples of, for example, I don't know, a firm with 10 CEOs all paid 1/10, the amount. Or no CEO, but we distributed the savings to other low level employees or whatever, right? So if the executive compensation consultant's job was not benchmarking and tax optimization and peer group measurement and stuff, but instead to identify the thing that will, the set of conditions that will get amazing performance, high energy, happy, joyful performance out of the CEO. And identify the incentives to get them there. And the metrics to make sure that they're behaving well. And to use that and advise the board on it to help them understand here's who we hire, here's how to activate them, and here's how it's going to benefit the organization. And here's how we tie that to incentives, and so on, so on. The other benefit of this is it sounds like a really fun job. I don't know, what do you think?
Matt Voiceover
Thanks for listening. Just to be transparent, part of my hope with this episode is that one or more of the executive compensation experts in my community will hear this and want to have a conversation about it. Not in the juvenile “debate me” sense, but more in the “what can we do together to make a positive difference in the world” sense. I suspect they’ll have a lot of insight as to why I’m kinda full of it, which would be helpful. It’s always good to find out that you’re wrong, so that you can be more right going forward. Still, if I'm being naive here or otherwise making mistakes, I think it's important to appreciate that none of our current executive compensation approaches are rooted in evidence. We have no idea if they work better than some completely different approach. To me, that means we have an amazing opportunity to experiment. If you're listening to this or reading the transcript and you have an opinion, please let me know you can email me at soundup@groundupgovernance.com. Let's see if together we can sort this out. Thanks again for tuning in to sound up governance. See you next time.
Sound-Up Governance (ep.42) - Executive compensation is probably broken (feat. Matt talking to himself)