ISSUES IN EXECUTIVE COMPENSATION

Speaker A Just a couple points. I've been negotiating employment agreements for executives for over 20 years across three incontinence. Unlike the compensation consultants, I have been the market maker. I know what the market is. And I think there are, I guess, a couple of opening comments. First of all, when you go into most companies, I'll make a very strong claim today. Most job evaluation systems and most corporations are fundamentally broken. I just returned from the world of work, which is the American Compensation Association's event 2000. People from around the world just put out their brand new book on how to set up pay grades. It's all broken. They're telling people best practice to set up pay grades is 15% midpoint differences. And all these overlaps when you create all those overlaps because what they're concerned about is delivering pay, not making pay work, and therefore the pay equitable. Then you end up with a fundamental problem and a disconnect between pay and work. And more importantly, what we see in a lot of organizations is as a result of a broken job evaluation system, poor titling and overlayering depending upon the size of the organization. It would not surprise me that you are probably looking at anywhere from 20 to $200 to $2 billion of excess compensation cost due to overlayer. And the HR guys are the ones who fall because they're putting forward a broken job evaluation system that does not link the work with the pay, results in overtightening, results in an organization structure that creates a whole host of organizational and people dysfunction that Elliot uncovered many, many years ago on the rest of the folks. And unfortunately, when you sit down with these 2000 people down in Orlando, Florida, as I did some eight weeks ago, they think what they're doing is working real fine. And I know Elliot had tried for years to crack the nut with these folks and didn't make any headway. We may start to make some headway now because what they're doing is going to probably get taken to court and get found out. So anyway, my opening claim was that.

Speaker B The overtling, et cetera, probably costing, depending upon your organization, anywhere from maybe 20 million to 200 million or more a year. And so if you're in compensation, you might want to take a hard look at it. And if you're not in compensation, you might want to blow it out. We have with a couple of our clients, nancy talked about the point here about determining the level of work. It is absolutely the starting point, not for the purpose of compensation, but for the purpose of getting the organization designed right, to get the work done. And for some reason organizations seem to see compensation and job evaluation over in this other pot over here. And I would suggest if you're doing it right, it's not a rework. It is the work of getting the organization structured right. And once you've done it, the pay automatically the pay structure, the pay of job analysis, job design system automatically just falls out because it's one same where it should be. Too many corporations, the managers are out designing the work structure, and then HR walks in and says, oh, let's give you the job evaluation system. The two aren't even lined up. I talked earlier today about thinking about levels of work process innovation, new product, new service market innovation, new business model innovation, global industry structure innovation. Over 15 years of research, I don't know how many client engagements we have arrived at least our point of view that to design the work and to build a job design structure and a job evaluation structure, we've actually ended up with about nine factors that we look at and have our clients look at, which include innovation includes Elliot's, time span includes the customer. Now, there's a scary thought, because at each level of work, maybe everybody should be doing something different for the customer. Are they the ones that write the check? And so we've arrived at a framework of looking at about nine factors of how to look at this from the resource complexity. What was Elliot measuring when he was measuring time span? As I alluded to both earlier today, go back and read measurement and responsibility. Time span is a great measure to Nancy's point. You got to make sure you're doing it right, and you got to make sure you're getting it from the right source. But it's a proxy, I would suggest, as Elliot did, it's a proxy measure. And it's partly a measure for resources under discretionary control, and it's partly a measure of getting to what level of innovation and creating what level of value for whom. And so I'll move on. But I just want to make the point that when you look at most of the job evaluation systems and I have all the books from World at Work from eight weeks ago, I've read them all, and they're all broken when I say they're broken. They were set up for an industrial era of capturing the work, depending on how I flip the chart over, but basically capturing operational work at work levels one, two, and three, and that's what they capture. So I was with Watson Wyatt in the United States a couple of months ago for one of our clients. I had one of my really smart guys with me. Happens to be an actuary by background. He does numbers for breakfast. You don't want to mess with those guys, believe me. Isn't that right, Nancy? Oh, yeah. And dinner and lutz. And he's been buying and selling companies for 20 years. Put a valuation on him. We talked about doing the math on getting the work right. Actually, let me digress for a second, because let me come back to the issue of what we are paying executives for. And the problem today in North America is what we're basically disclosing in proxy statements for shareholders is basically operational work. So it really raises the fundamental question, which is why the institutional investors are asking me to come talk to them that, are we paying CEOs for the right work? And based on what we're seeing in a lot of the proxy disclosures, the answer is no. And beyond that, we're overpaying them for the operational work they're doing. That's a quick digress. Anyways, back to our trip to Dallas. Nancy went off to Philadelphia. The only two folks that got to the Dallas ice storm were the two Canadians. The folks who live in Dallas actually stayed home and called in on the conference call, as we said in Watson Wyatt's office. Anyway, the quick story I wanted to tell you is, I said to the folks at Watson Wyatt and we made it, but we know nothing about compensation because they were really uptight. We're just organizational consultants, and we're helping this client out. And so just want to understand what you do and how to do it to help us with our client, who you guys have been advising for like, 15 years or whatever. I said I want to draw this imaginary line, and below this imaginary line are because you guys go out and do all this benchmarking for our client, right? And you put together this pay report. The pay report says, oh, our client's being paid at the 40th percentile, so therefore everybody gets a raise or should get a raise because we're overpaying them. We're underpaying them. That's a board decision or management decision. I said, well, just in interpreting this pay data, just so I'm clear, because I'm an organizational guy. I don't know anything about this compensation stuff. If we looked at CEO roles, COO roles, chief Financial Officer roles, chief marketing roles, and those roles who are, I'm going to draw an imaginary line think about this that are primarily doing that work just to run today's business. Now, above that imaginary line are the same roles title. Kind of a crazy job description in your book that you've given me here, and I have this pay data you've given me from Mercer and Watson, Wyatt and Tower's parent. I'm looking at all this data. I said, to what extent does Mercer and Watson and Tower and yourselves and all this pay data I'm looking at, would it reflect work that might be more looking, developing new products and developing new businesses, possibly even developing new industries? And they said, we don't do that. Oh, okay. Just want to be clear. So what they basically told us, we didn't make a big noise about it is that all of the pay data you get today from Watson and Towers and Perrin and Mercer, maybe that's why you decided to leave. That's why everybody's leaving Mercer anyways, is that they do not. What they basically do is the pay data that's here at these possibly different levels, they've taken it all, and thrown it into a big pot. And as Kenwitz say, they've created more statistical noise in that data than you can shake a stick at. Look at it this way. They have these surveys. You now go on the Internet. Now if you're the compensation manager for Procter Gamble and the compensation manager for Campbell's, and you go to the survey site for Mercer and you look up and say, yeah, I got a CDO, I got a Chief Marketing Officer. Yeah, I got a Chief Financial Officer, and you put your pay data in. But I've been with the board members. I mean, if I said, AG, laughley at Procter Gamble, AG, do you think I'm going to get you to come leave your job at Procter Gamble and come take on the job at Campbell Soup? He'd laugh at me. Why? It's about a five factor difference in complexity of the real work, and he knows it. I haven't seen him about four years, but I think he'd probably give me that answer. If you compare Eli Lilly, who's in the pharmaceutical business, to Johnson, and Johnson, who is in six sectors of businesses in 57 countries around the world with their own venture capital company with 400 million in capital to create what new business models? Do we really think the CEO of Eli Lilly's role is the same as the CEO of Johnson Johnson? I don't think so. But they're putting all that pay data into those tables and giving you a report and saying, there's, the number doesn't work. A couple of them are going to get sued in the next little while, and we'll see what happens on that. Let me see what else we're going to talk about. Oh, we had a fun trip to Watson Wine, so that was good. Anyways, what we did do, two clients in the last year were they had existing contracts with these traditional comp consultants. So we could they were used to doing in some cases at one client it was 200. Another client, it was like 600 jobs a year where they would go out and what, call market price, every job. Well, that keeps a compensation department of 510 20 busy just doing nothing. But this thing called market pricing doesn't help the managers. So for this one client, they said, well, we can't just totally throw it out. Can you help us figure out how to get there? And we said, okay, we'll create a bridge. So when we initially went through this one process, we went and did job matching. We took 20 jobs in what we call non scarce roles. So it was HR accounting, it was HR Accounting, Finance, Marketing, and the call centers for this one client. And we went out to the Watson Mercer job books where you do this job matching idol. We found the pay, and then we did the same thing at level two. By the time we got finished, we had absolutely no confidence we had less than a 50% confidence level in the pay data at three to help us try to get to X. So we threw it all out, all of it. Because when we tried to use it, it totally distorted the pay numbers we were getting and trying to get to X. Couldn't get there. Did a couple of runs, went back to he says, that seems real. Then Ken, my analytical guru, goes, well, what's your probability constant level of the numbers? And you've got the job match high, medium and low in three, right? This is pretty low. He says, Then get rid of the data. We did. What we ended up doing is taking those same job families at level two only for which we did have a high comfort that it was either a mid or high two. For the mid two, we did some statistical calculations to approximate high two. Well, three. Put it all together and we had a small statistical sample, 4600 jobs in those four job families. And what did the number come out to? $113,000 for total direct compensation at high two, bottom of three. It sounds like somebody got about the same number in Calgary. I'm glad to hear that. We took that number, popped it into our Felt Fair pay table and dove six steps, one sheet, give it to the managers, we're done. Can't trust any of that data coming out of Watson, Wyatt Towers. And Parent and I'll go to court with them any day because we know where all the dead bodies are buried in their process. We sat there and had them explain it to us. So anyways, lots of challenges there. And to kind of complete the circle, we did know in the United States, there's this real problem right now in executive pay seeming somewhat off the rails. And when you ask anybody about this thing called self fair pay, everybody goes, Never heard of it. When you sit at a conference with the supposed thought leaders in executive compensation in America, nobody's heard of it. When I put some of that up, people go, that's fascinating. I said, well, you know, there's a whole bunch of research studies on this about 2030 years ago. Oh, really? Yeah. So what we decided to do is we took America apart. We took the Russell 3000, scrubbed the data, ended up with about 2500 companies. And of the 2500 to 2000 companies remaining for which we had three year data, we just went for fun and giggles to go take a look at what was the pay differential between the CEO and the named executive officers at the next level for all 3000 publicly listed companies, including data on you. Throw the outliers out. And there's a good news story. The good news story is that the pay multiplier came up to 2.4 research, Roy Richardson's research, and Edahoma's research, all a long time ago that this thing, this pay multiplier Felt Fair was maybe in a two times range. 2.4 didn't seem that bad. That was the good news story. The not so good news story is we had 500 companies, many of them in the Fortune 500. Fortune 1000 were the pay differentials between the CEO and the next level down. Was it 1020, 30, 40 and 50 times, just the next level of management. And you kind of go, how did they ever get there? That's because nobody ever decided that internal Felt Fair pay was worth looking at. So the good news is that the felt fair pay model and the two times multiplier when the 27 variations of the spreadsheet nancy, because we got all these rocket scientists on our team, we can plug in. You want fell fair pay at 2.1? I'll give it to you. You want 2.5? I'll show you what it looks like, too. The interesting part, with two of our clients, we went back and reran the numbers even at two times, and we put it against their current pay. But at level four, what do we have? 100% fit. Level three, what? We have 100% fit to current pay distributions in terms of the structure, just current pay. Level two, we had about 60% fit, and at level one, we had like 45% fit. A whole bunch of reasons for that you can choose. So the good news is this concept of welfare pay, the concept of a pay multiplier to help you get a truly differential work, seems to work. I just negotiated another employment agreement last week for an executive vice president. I looked at what level of work it is, roughly. I looked at even the 113 pay multiplier and looked at what we're bringing them in at and said, yeah, it's feeling about right. So a couple of our board members are wanting to throw Watson, Wyatt and Mercer and these guys out, but they're looking for another way to do it, and this is it. So when you bring in market competitive pay at level two and take a welfare pay up vertically, as long as get you to some level of a structure, and we did exactly what you did. We took a TDC number, total direct compensation number, and then we reverse engineered back from just as Dom did, to, say, make a decision on what that split you'd like to be between long term. And what we did was three of our clients LTIP was at level four and five or six. We provided LTIP as well. For one client at top of three, the bonus was at three, short term center plan. So the fours and fives, as an example, would be getting both climate to base. And then at levels one and two, it's just straight comp, no bonus. And we have one organization, they get like, I don't know, 500,000 employees at that level of which, as Ken would say, it's a 4% incentive kind of directionally, insignificant or insufficient, as he would say to motivate any change of behavior. Now we're going to be challenged to convince them that maybe to administrate that for 1000 employees ain't worth the juice and the squeeze of what they're going to get in any change in motivation. But basically levels one and two, as Elliot would say, just pay them for the work at level three, because we think that is the clear processed owners of the current business and link it to the year over year over year dollar value creation. I've convinced one of my board members, none of this budget to actual target stuff. So you sandbag them on the budgets and you get the budget to actual target and you go bam, they actually created real value year over year over year. We'll give them a percentage of it at least. Our board, we just suggest going that route versus no, here's our budget, here's our target, we exceeded it, we sandbagged you on the targets and now you pay us. Did you get any value for the customer? Did you create value for the shipper? No. So we kind of want to move away from that. That's a long route in terms of combination of internet. Get the work right, use the right market comparables, I would say you can use some of the survey data, just be very careful. I threw more job matches out for three clients in the last twelve months. And I looked at the HR folks sitting in the room with me and I said, who would prove this? The CEO. I said, well he's wrong, throw it out. I can't take it to the board of directors. I'd get shot if I ever went into court with that job match. You don't have a job like that. We would really end up I guess our one key learning for two clients in the last 1218 months is at three, I wouldn't trust the data with like a stick. Literally you cannot figure out between the job descriptions and the titles that they're using, whether it's high, medium or low, even if you wanted to. And so when you put it in, we're running statistical observations in huge databases and it totally throws it out. So we literally in the end, after number of runs at it, just said get rid of it. And that got us cleaner data. We're feeling 90%. And then to Nancy's point, the other thing on the market comparable, every time you interview somebody, write down what they're really supposedly earning. Every time you lose somebody, where'd they go and how much did they go for. The problem with all of these surveys is they're attempting to get at people who are in place. But how many people make a lateral move for no change? I've been doing this for 20 years. It's rarely I can get somebody to make a lateral for no compensation change. But to Nancy's point, to Mr. Morgan's point, Rich, I mean that's why we're talking one industry, we're talking salary, which is the second largest data provider in the United States, to actually go out in this one industry sector. And we're going to redo the whole industry for 200 companies, because that's the only way we can come up with maybe real data that we got trust in, that I can go to court and defend if one of the shareholders comes in and wants to take somebody over ten. And then, as Vancy said, determine the X. We got the X, and we came up with 113. I got 113, he's got his 96, or whatever. It depends on the organization, where it shows up, and then pop it in the fell for a pay table and do exactly what Don did in terms of a policy decision. It's really that simple.