Shareholders or stakeholders: three answers for managers

On 20 April 2010 the Deepwater Horizon rig in the Gulf of Mexico exploded and 11 people died (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. vi). The US national commission that investigated wrote that many decisions increased risk while saving significant time and money, and that the BP team at Macondo appears to have had no formal system to prove the alternatives equally safe (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 125). The question for any manager: when saving money cuts against safety, whom do you owe first, shareholders or everyone affected?

Where do the three answers differ?

The library's Shareholder, stakeholder and market-failures views framework sets them side by side. What follows is the library's version, with nothing added by me.

What does Friedman say, including the parts easily overlooked?

Friedman (1970) argues from the agency relationship. The manager is the agent of the owners, whose direct responsibility is to run the company as they wish, generally to make as much money as possible, within the basic rules of society, both law and ethical custom (Friedman, 1970). If a manager spends on social ends on his own initiative, he writes, that person is spending other people's money, whether the shareholders', customers' or employees', and doing the government's job of taxing and spending (Friedman, 1970).

Three parts are easy to overlook, in my view. First, he does not forbid conduct called socially responsible when it serves the company's long-run interest. His example is a major employer in a small town spending on local amenities to attract staff, which he treats as a legitimate business expense. He observes that, in practice, "social responsibility" is frequently a cloak for actions justified on other grounds (Friedman, 1970). Second, he sets the individual proprietor apart, who spends his own money (Friedman, 1970). Third, his closing line carries a condition, that profit comes with open and free competition, without deception or fraud (Friedman, 1970).

The weakness the library framework records: the argument overlooks other people's rights where the law is weak.

How does Freeman change the question?

In my own words, Freeman (1984) changes the question from "whom does the company belong to" to "who does the company touch". The library's concept of stakeholder theory, taken from Freeman, sums it up: managers must balance the rights and interests of every group the firm affects, not only shareholders (Freeman, 1984).

The framework gives the core reason as each group's right not to be treated merely as a means. It also names the weakness: a manager can invoke "stakeholders" to escape accountability. If you take account of every group and nobody scores you, any choice can be defended.

Where does Heath stand between the two?

According to the library framework, Heath (2014) keeps the goal of maximising profit but forbids exploiting market failures. The library's concept entry lists them as weak competition, asymmetric information, externalities and public goods. Also per the framework, institutions divide the work: the market makes the pie bigger, the state shares it out, and the manager simply may not earn from places where the market is broken (Heath, 2014). An externality is a cost that falls on a third party to the transaction. The detail of that argument sits in the post on misleading advertising and broken markets, so here I use it only as a lens.

The framework records the weakness as exposure to the "second best" objection (Heath, 2014); I have not read Heath's book, so I do not set that objection out in detail. The post on when profit is legitimate goes further into the market-failures approach.

What does the report on Macondo record?

The first four points are what the commission wrote, with report pages; the last comes from the Department of Justice.

• The available evidence does not show that many decisions went through formal risk analysis. The commission wrote that decision-making at Macondo did not ensure personnel fully considered the risks of time- and money-saving decisions. Whether purposeful or not, many risk-increasing decisions clearly saved the companies significant time and money (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 125).
• Choosing a cheaper option is not wrong in itself, provided it is proven equally safe. The problem was that the BP team at Macondo appears to have had no formal system to check this (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 125).
• The commission cannot say whether anyone consciously chose a riskier option because it cost less (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 125).
• It called the root causes systemic, not the product of aberrational decisions by a few individuals, and traced most of the errors to a failure of management, together with failures of government oversight (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 90, 122).
• The consequences: BP had to pay out tens of billions of dollars (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 89), and according to the US Department of Justice's release of 15 November 2012, BP (through its subsidiary BP Exploration and Production Inc.) agreed to plead guilty to 14 criminal counts, 11 of them manslaughter, with $4 billion in criminal fines and penalties (US Department of Justice, 2012).

What do the three lenses see when placed on this case?

This part is my inference. The authors did not write about BP.

The three "what it asks for" cells meet at what the commission asked: when procedures change there must be at least some formal risk analysis, and if time is short only proven alternatives should be considered, and companies need strict policies requiring rigorous analysis and proof that a cheaper alternative is equally safe (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 126). All three lenses support that step, each for a different reason. For Friedman it protects owners from risks they cannot see. For Freeman it gives a voice back to the affected group. For Heath it blocks the pushing of costs outward.

They differ when the law is silent. Friedman stops at law and custom, while Freeman and Heath ask on behalf of those not in the room. They also differ on who balances conflicting claims: the democratic government (Friedman), managers too when the state fails (Freeman), or an institutional division of labour between market and state (Heath).

What three questions should a manager ask before deciding?

This is my own set, combined from the three lenses and untested.

1. Who loses if I am wrong, and do they have a voice in this meeting?
2. Which law or standard already binds this decision, and who has checked that the cheaper option still sits within it?
3. Does the saving depend on someone outside the transaction bearing a cost that I do not pay?

If none of the three has an answer, then following the Macondo case, do what the commission asked: analyse rigorously and prove that the cheaper option is equally safe before changing anything.

Going deeper: what I read, and what criticism is missing

I read Friedman (1970) in full from a reprint of the article in the New York Times Magazine. That copy carries no magazine page numbers, so I paraphrase rather than quote with pages (Friedman, 1970). Freeman (1984) and Heath (2014) are books I have not read. What I say about Freeman rests on the library's concept entry, and what I say about Heath and the weaknesses on the library framework's notes (Freeman, 1984; Heath, 2014). So this post does not criticise the two books in depth, and the weaknesses above are the ones the library framework records.

The commission's report is very long. I read the preface and the passages at pages vi, 89 to 90, 122 to 126 and 167, plus a footnote on page 346, and not the rest. Over four million barrels spilled is the government's estimate (about 4.9 million), and BP disputes those figures in a footnote of the report (National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, 2011, p. 167, 346). The Department of Justice page blocks automated access, so I checked the content of the release through a reprint (US Department of Justice, 2012). Note too that the commission's report and BP's criminal counts are different kinds of evidence: one is an investigation, the other a plea agreement.