The Original Incentive Plans of General Motors and Du Pont
In 1920, over one hundred years ago, Alfred Sloan inaugurated the decentralization of General Motors. In 1921, Pierre S. du Pont followed suit at his namesake company—not terribly surprising given Du Pont’s 30% ownership stake in General Motors. Both companies began a quest for a multidivisional form of organization to better handle the size of their respective operations.
In the case of GM, it was already quite a large operation in 1920. The company then employed 97,376 people at the height of that manufacturing season. It had achieved record annual revenues of $567 million. Net earnings available for dividends were $37.75 million on tangible assets of $582 million. GM also had over a dozen divisions and owned many other related businesses. The timing seemed right to make the switch to decentralization to better manage for the continuing growth ahead.
And GM continued to grow—it nearly doubled in the seven years after 1920. In 1927, GM employed $900 million in gross assets and 160,000 employees across numerous operating divisions. Interestingly, GM also engaged in the habit of enunciating its management ethos and procedures for all who cared to hear or read. For example, Donaldson Brown, a Vice-President at General Motors read a paper before the American Management Association in New York City entitled “Decentralized Operations and Responsibilities With Coordinated Control”. Brown was there to explain why and how GM managed its vast collection of assets.
Brown explained that GM management had ultimately recognized the limits of a heavy, top-down management style. GM management also knew that continuing to relieve people of responsibility would also rob them of initiative. Conversely, if initiative and hard work was the desired result, this could be best achieved by delegating as much responsibility and authority as possible.
Another benefit of decentralization and the creation of many distinct operating units is that it is easier to gauge the effectiveness of management. Brown wrote (emphasis mine):
“In gauging the effectiveness of management the first approach always is to examine the over-all result—the rate of return enjoyed on capital employed. If this be sub-normal, having due regard to the character of business and the competitive situation, it is self-evident that something is wrong. The second step is to identify the cause. With the segregation of General Motors into separate businesses of distinct classification we are in position to compare the performance of various divisions respectively with competition in the same line. Thus we can gauge the overall effectiveness of management in each case.”
And what was the primary yardstick for measuring the efficiency? Not margins. Not sales growth. For Brown, Sloan, and the rest of GM upper management, the yardstick was return on capital employed.
“The general test of efficiency of management of any business is the rate of return on capital employed. Capital in industry is entitled to a varying rate of return largely dependent upon competitive conditions in the broad sense and the hazard that is involved. Needless to say goodwill, that intangible and illusive, but none the less valuable, asset which every business seeks to enjoy, is of great importance.”
Compensation and Incentivization
Given that GM and Du Pont embraced a decentralized approach that endowed great freedom and responsibility to its division leaders, the next step was to figure out a way to incentivize those people to act like owner managers. With both companies having a shared interest in each other, both developed management incentive schemes that were nearly identical—and quite brilliant for the time. Du Pont introduced its scheme in 1927 and GM its scheme in 1930.
For the Du Pont scheme, an Executive Trust Plan was created. Select managers were invited to participate. The Executive Trust Plan then purchased common shares from Du Pont at market prices using a mix of cash and debt. Part of the cash came from the bonuses of the participating managers, which in turn came from 6% of the amount by which Du Pont’s earnings exceeded a 6% return on capital employed. For GM, this bonus calculation was 5% of the amount by which GM’s earnings exceeded a 7% return on capital employed.
The debt that was used to fund the purchase of Du Pont stock for the Executives’ Trust came in the form of notes at market rates of interest with 7-10 year maturities. This debt had another interesting feature that Richard Holden describes in his paper “The Original Management Incentive Schemes”:
The notes were not cash-pay, rather pay-in-kind—that is, executives were not forced to make cash payments and could allow the interest to compound. The interest could, however, be paid down by dividends on the stock, as well as by bonuses received under the “B” Bonus plan and the additional bonuses provided under the Executives’ Trust Plan.”
Holden also provides a nice visualization of the incentive scheme in the same paper:
The Most Amazing Thing
As Holden further wrote, “the analog of the Du Pont Executive Trust Plan at GM was the General Motors Management Corporation (GMMC),” which was organized in 1930, five months after the stock market crash of 1929. Right as the Great Depression was beginning.
In 1930, GMMC purchased 1.375 million shares of GM common stock at $40 per share. This consisted of $5 million in cash and $50 million of 6% notes. The plan was for dividends on the stock and the individual bonuses of the executives to redeem the debt.
But the price of the GM stock fell. Two years after the compensation plan started, GM stock traded at $8 per share. GM was not meeting its return on capital employed threshold for executives to earn their bonuses. GM had also slashed its dividend from $3 per share to $1 per share. GMMC was unable to pay the interest on the notes used to acquire GM stock just a few years earlier. Over the next year or two, there was a small battle between management and the board of directors on how to resolve the compensation scheme in the face of the serious issue of the decline in share price and dividend per share.
Sloan and other managers favored setting the entire plan aside and replacing it with a new plan. According to Richard Holden, several GM board members emphatically refused this course of action (emphasis mine):
“In particular, two board members, George F. Baker Jr. (chairman of the First National Bank of New York) and George Whitney (a partner at J.P. Morgan and Company) saw the problem as one between a creditor and a delinquent debtor (Cheape, 1995). At Sloan’s behest, Walter Carpenter attempted to negotiate a settlement, which would have included GM agreeing to buy back the shares at the original $40. Baker and Whitney would have none of this, and Baker, especially, insisted that any change be approved by the stockholders. Indeed, Baker was seen by Whitney as being rather lenient in not pressing for liquidation of the scheme and hence management forfeiting all bonuses earned since inception (which amounted to 170,000 shares trading at about $28 each at that time) (Cheape, 1995). The board agreed to defer shortfalls of interest payment and principal, but steadfastly refused to guarantee a buyback at $40.”
I think it is amazing that the GM board stood by the principles that were the bedrock of the incentive scheme in the first place. Management had the opportunity to participate in the upside and the downside of the business. Removing the downside would have been counter to those principles. Without the downside, the scheme is just a bastardized version of a salary.
However, GM’s share price would eventually recover beginning in 1934. Profits eventually increased again to the point that it allowed for bonuses. And Holden shares in his paper what I think is the most amazing thing:
“The [incentive compensation] plan ended on March 15, 1937, with GM stock standing at $65 per share. The final return for each $1,000 originally invested was $12,595…. This included bonuses and dividend, less interest paid on the borrowed funds, and represented a return of 43.6 percent per annum over a period of seven years, during the Great Depression.”
If anything, this example just goes to show that the waiting is one of the hardest parts about investing and business.
In Summary
Although I’ve not studied every single management compensation scheme that existed a hundred years ago, I have looked at hundreds of proxy statements of companies over the last fifteen years. There are usually two components to modern day schemes. One typical component is the gifting of options to management that vest over a period of three years. The second component is some level of restricted stock units that can be earned by meeting some sort of financial metric (growth in sales, growth in EBITDA, growth in EPS, etc.) over a year or a period of years (usually never more than three).
Most modern day incentive schemes pale in comparison to what GM and Du Pont offered and expected from their managers. The modern day schemes rarely require management to have skin in the game by requiring them to purchase shares of common stock—at market prices with their own money—and then wait for seven to ten years to vest.
Two standouts of current public companies, in terms of compensation schemes, are TransDigm and Constellation Software. According to TransDigm co-founder Nick Howley, TransDigm to this day still underpays its managers with salaries and over-equitizes them with options. But these are truly unique options in the world of public companies. They are 100% based on performance—there’s no time vesting, no “pulse vesting”, no ability to have those options vest simply being alive. For options to vest to a TransDigm executive, the annualized return on intrinsic value of TransDigm must hit a 17.5% hurdle. Constellation has a performance based comp scheme where if group operating managers meet certain criteria based on net revenue growth and return on invested capital, they are then required to use 75% of that cash bonus to purchase shares of Constellation in the open market, which are then held in escrow for a minimum average period of four years. TransDigm and Constellation are exemplars of forcing high standards and “skin in the game” on their executives.
As Holden suggests at the end of his paper, for most other public companies, it is likely worthwhile to review the 100-year-old management incentive schemes of GM and Du Pont. Longer-term thinking (and maybe performance) could be vastly improved.
Please Subscribe
If you enjoyed this content, please share and subscribe.
Sources and Further Reading
Brown, Donaldson. “Decentralized Operations and Responsibilities with Coordinated Control”, General Motors Corporation, February, 1927.
Hoover, Gary. “The Greatest Businessman in American History: Alfred P. Sloan, Jr.”, Archbridge Institute, October 29, 2021.
Holden, Richard T. “The Original Management Incentive Schemes”, Journal of Economic Perspectives. Volume 19, Number 4, Fall 2005. Pages 135–144.
Constellation Software, Inc. 2024 Management Information Circular. Pages 12-13.
Disclaimers
The content of this publication is for entertainment and educational purposes only and should not be considered a recommendation to buy or sell any particular security. The opinions expressed herein are those of Douglas Ott in his personal capacity and are subject to change without notice. Consider the investment objectives, risks, and expenses before investing.
Investment strategies managed by Andvari Associates LLC, Doug’s employer, may have a position in the securities or assets discussed in any of its writings. Doug himself may have a position in the securities or assets discussed in any of his writings. Securities mentioned may not be representative of Andvari’s or Doug’s current or future investments. Andvari or Doug may re-evaluate their holdings in any mentioned securities and may buy, sell or cover certain positions without notice.
Data sources for all charts come from SEC filings, Koyfin, and other publicly available information.