Larry Ellison, Model Executive
Larry Ellison gets paid a lot. He owns one quarter of Oracle, the software company he founded in 1977, yet the company’s board has kept awarding him millions of options to buy Oracle stock every year.
But hey, Bloomberg News’s Caleb Melby reported Wednesday, he's a bargain!
Oracle’s total return has trailed the Standard & Poor’s 500 Index over the last five fiscal years ending in May, a performance that prompted the software maker’s shareholders to majority-vote against its pay plan three years in a row, even though about 25 percent of the votes are Ellison’s. With a fortune approaching $45 billion, the 70-year-old has received 7 million options every year from 2007 to 2013.
It turns out Ellison may be a pretty good deal. While Oracle’s stock lagged, a different measure of performance -- economic profit, which is defined as after-tax operating profit minus the cost of capital -- shows the company has on average outpaced 98 percent of companies in the S&P 500 over the past three years.
The debate here is really over the metric that should be used to judge a company’s performance: total return to shareholders or economic profit, which in addition to the items noted above counts R&D and employee training as investments, not expenses.
Total return to shareholders, net income, free cash flow, economic profit, adjusted OIBDA -- they all have their strengths and weaknesses. Over a horizon of 15 years or more, total return is perhaps the one metric to rule them all. But in trying to predict long-term total return, you’ll often get more information from economic profit than from total return or any other measure.
This was the argument of Alfred Rappaport, the Northwestern University accounting professor emeritus who coined the term “shareholder value.” In the 1980s his Alcar Group, along with Joel Stern's and Bennett Stewart’s Stern Stewart, proselytized that economic-profit-like measures were much better than earnings per share at identifying the drivers of long-run value. The idea that, say, three-year return to shareholders was the right metric by which to judge a corporation’s management was not even on the table at the time.
In the 1990s, though, the term “shareholder value” -- much to Rappaport’s chagrin -- came to basically mean, “What’s your share price today?” And as regulators and corporate governance watchdogs tried to get a grip on exploding executive paychecks, total return became the standard performance metric.
U.S. law now requires companies to report their five-year shareholder return relative to an appropriate benchmark in their annual report or proxy statement. When shareholder-advisory companies, including Institutional Shareholder Services and Glass Lewis, decide whether to recommend a yes or no vote on an executive-pay package, the alignment between pay and total shareholder return is often the deciding factor.
By that measure, Ellison’s pay package has been an outrage for at least the last five years. Judged against economic profit instead, it looks … average. To quote again from Melby’s article:
Ellison’s $103 million pay for fiscal 2014 equates to 2.1 percent of Oracle’s three-year average economic profit, which is better than 52 of the CEOs at companies in the S&P 100 index, according to data compiled by Bloomberg.
The best economic-profit-to-pay ratio in the S&P 100 belongs to Google’s Larry Page, who gets paid $1 a year. Other high-profile CEO/founders, such as Facebook’s Mark Zuckerberg, Berkshire Hathaway’s Warren Buffett and Microsoft’s Bill Gates back in the day, have also drawn nominal paychecks and received no stock awards. John D. Rockefeller handled things that way, too (although Standard Oil paid huge dividends, so he wasn’t exactly starving himself).
Ellison most decidedly does not follow this practice. One long-popular theory for why he let/made the Oracle board’s compensation committee keep piling on the option awards was that he wanted a better shot at catching up with Gates to become the world’s richest man. Gates's net worth is currently about $83.1 billion and Ellison's is $44.9 billion, according to Bloomberg’s billionaires list, so that seems like a lost cause.
A less-pathetic explanation is that Ellison wanted to make sure he maintained control of the company he founded. Unlike Google, Facebook and Berkshire Hathaway, Oracle doesn’t have a dual-class share structure, so keeping his stake above 25 percent -- even as Oracle bought other companies -- may be a strategic imperative.
As of this year, Ellison's options gusher is beginning to subside. The company’s new co-CEOs, Safra Catz and Mark Hurd, will each get more options and performance stock units in fiscal 2015 than its new executive chairman and chief technology officer (Ellison).
In the end, the question of whether it’s really appropriate to shovel so much money and shares at Ellison and other top executives probably can’t be answered purely with reference to measures of corporate performance. Morals and politics have to come into play, too.
I also think, though, that corporations should be able to experiment with different methods of governance and executive compensation and not have to adhere to a strict set of best practices handed down by ISS or Glass Lewis -- or, for that matter, by Congress or the SEC. Oracle is a good case in point there. It gets lots of flack for its governance and pay practices and yet, as I have written before, it seems to be pretty well-run.
How do I measure this? Well, first by the elegant brevity of Ellison’s executive bio on the Oracle website, which reads, in its entirety:
Larry Ellison is Executive Chairman of Oracle Corporation and Chief Technology Officer. He was CEO of Oracle from 1977, when he founded the company, until September 2014. He also races sailboats, flies planes, and plays tennis and guitar.
And then there’s a more serious metric. Total return to shareholders, but over 28 years (Oracle went public in 1986):
Update: A reader wondered about this, so I should clarify that the above chart isn't meant to show the total return of Oracle and Microsoft since their initial public offerings. Oracle’s IPO was on March 12, 1986 and Microsoft’s on March 13, but for some reason the magical machine I use to calculate total return doesn’t have the first month of Oracle’s stock data, so I started my chart at the end of 1986.
Microsoft had a significantly better 1986 than Oracle -- its stock price rose 127% from its IPO to the end of the year, while Oracle’s was up 38%. By my back-of-the-spreadsheet calculations, Oracle’s total return since its IPO still beats Microsoft’s, although the margin isn’t as dramatic as in my chart. Also, while I followed current Oracle practice in referring to Ellison as “founder,” that means he’s A founder, not THE founder. There were two others.
(Updates explanation of third chart in final paragraph of article published March 12.)
I'm not sure what the strengths of adjusted OIBDA are, but I'm sure some media industry chief financial officer could enlighten me.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story:
Justin Fox at firstname.lastname@example.org
To contact the editor responsible for this story:
Paula Dwyer at email@example.com