Duly Noted: Solving the Principal-Agent Problem in Firms:The dumbest idea in the world?

This article in Forbes argues that a new book by the Dean of Rotman School provides an antidote to the rampant excesses of modern day capitalism.  The principle swipe is against the landmark paper (over 29000 Google Scholar citations)  by Jensen and Meckling on both the prevalence of the principal agent problem in the governance of firms and the various solutions to overcome it – including creating incentives that maximize shareholder value.  Quoting Jack Welch, former CEO of GE, the article says that maximizing shareholder value is the dumbest idea in the world.  I my self am not sure if this is THE dumbest idea in the world – in fact there are many more that would easily surpass P-A problem resolution – but I am sure this will ignite a debate about why firm’s exist – what is the best governance mechanism for them and the role of economic theory and action in our lives.  I for one need to go back and read the article and then read the book.

6 Comments on “Duly Noted: Solving the Principal-Agent Problem in Firms:The dumbest idea in the world?”

  1. Indy says:

    I think an interesting comparison is to graph GE on Google Finance for (All) – the option longer than 10 years, which gives you from 1978. You can then compare it to the S&P 500.

    Around 1996, GE really takes off and over performs, until around 2009, when it seems to abuse a common phrase “revert to the mean.”

    So if you bought in 1996 or before and sold after 1996, but before 2009, then relative to other investments, Jack Welch’s shareholder value concentration worked for you.

    I guess the question could be, who does 13 years trends like this benefit? And is that benefit valuable for society as a whole.

  2. srp says:

    The internal contradiction in the article (and, I surmise, the book) is that it claims that not focusing on shareholder value is better for shareholders–not maximizing shareholder value maximizes shareholder value. This is how Captain Kirk blew up all the evil computers on Star Trek. A greedy stock-incentivized CEO should, according to the argument, not try to maximize the value of his shares and instead do whatever virtuous thing the author believes actually improves the company. So then the problem must not be greed but systematic error and the compensation system is fine, destroying the thesis of the book. This is the kind of problem you should identify in the first five minutes of outlining your argument.

    The shoddy empirics in the article are not too encouraging either. Correlation doesn’t imply causality, and in this case we have a trove of evidence suggesting that the old non-stock compensation schemes led to featherbedding, empire building, earnings management, and complacency in the face of foreign competition.

    The control theory problem of what data to use in adjusting policy is an interesting one, and overreaction to swings in the stock price may well be a problem, just as overreaction to swings in machine efficiency can be a problem (hence the need for statistical process control). But that’s not an incentive problem.

  3. Todd Dunn says:

    This is a fascinating article. I agree with you Karim that this may not be the dumbest idea in the world, I read into this another way of looking. While today CEO’s spend so much time trying to set and manage expectations, I see it as a diversion from focusing on customers through figuring out how to innovate with employees, supply chain partners, customers, and even crowd sourcing. This “qtrly” expectation causes a significant conflict with creating innovative companies due to the incompatibility it creates between the FP&A cycle and innovation. That, to me, is the current danger of the way Wall Street expectations are managed. So many decisions are made with the short-term expectations focus versus considering the type of company that is being created or that needs to be created. It certainly causes companies to hesitate in making bold decisions, especially regarding how to innovate as you know so well!

    All my best!



  4. Joel West says:

    I haven’t seen Martin’s book, but Denning’s summary (or interpretation) seems to conflate three different issues: focus on the stock price, managing expectations and short-term vs. long-term orientation. While managing expectations is a new issue, the other two are old complaints that predate Jack Welch. (NB: US auto companies in the 1970s).

    Perhaps because he’s across the 49th parallel, Martin missed one key factor: Section 162(m) of the US Internal Revenue Code enacted in 1993. Intended to curb high executive salaries (by preventing tax deductions for salaries beyond $1m) the unintended consequence was to create incentive-based compensation anchored to low expectations.

    In the end, the complaints of Martin are the same as those of 10,20 or 30 years ago: the agents act in their own short-term interests while employees, communities and (some) shareholders have long-term time horizons. His solutions are radical and new but there’s no evidence they will be any more successful than §162(m).

    If GE stock tanked after Jack Welch left, should some of his compensation be tied to performance after he left? And how much of that is because his sucessor wasn’t as good (either in performance or in projecting confidence)? For four years, GE stock lagged the S&P and DJ by about 15%, about where it was a year after Welch left.

    Everyone Steve Jobs is what a CEO should be: he ran Apple for the sake of customers and Apple’s long-term future and shareholders won handsomely. But how many CEOs are as good as Jobs, and how many will do that good a job as a hobby? The majority of Jobs’ wealth came from buying Pixar cheap and selling it for 7% of Disney. How many CEOs would serve as CEO of one company for $1/year and director of another for $0 (instead of $236k/year).

    But for every Steve Jobs there’s at least another Ken Olsen or Scott McNealy who follows his long-term vision right into the side of a hill. Or then there’s Steve Ballmer, who presided over a 50% decline (after the bubble) followed by a decade going exactly sideways. No one would argue that these execs were not interested in the long term, but which of Martin’s reforms would solve these problems?

  5. srp says:

    Good examples, Joel.

    Other CEO “long-term” follies for the anecdote hopper: Richard Ferris’s Allegis plan to vertically integrate United Airlines with Avis Rent-a-Car and a hotel chain; Avery Seward planning ahead for the coming postwar recession that never came by starving Montgomery Ward of capital investment over a decade-long boom; Peter Grace building long-term positions in several awful industries without identifying any competitive advantage for W.R. Grace; the entire U.S. steel industry investing huge amounts in new long-term capacity in Iron ore mines and pelletizing facilities just as minimills and foreign competition were about to make all that capacity redundant; Roger Smith’s multi-gigabuck robotics investments to create a Brave New GM; and lots of great examples for the industrious to ferret out from Japanese companies in the 1980s and 1990.

    I think it’s an interesting question why critics focus solely on alleged short-termism when there’s also so much evidence of excessive long-termism. Is there any systematic evidence that one type of error is more prevalent or more damaging?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s