"Peer Benchmarking," the "Lake Wobegone Effect," and Outsize Executive Compensation: the Amgen Example

We recently discussed why health care's 99 percent may have reason to be upset with the one percent, the elite that run health care organizations. Once cross-posted on the KevinMD blog, it caused an eruption of controversy. This included some anonymous commentators who insisted that top leaders' made contributions to health care that were so heroic that their compensation was insufficient.

So conveniently an article this month in the Washington Post dissected the mechanisms that let pay of top health care leaders inexorably rise whatever their and their organizations' performance. 

The Amgen CEOs Compensation

It used as an example the compensation given to Kevin W Sharer, the CEO of Amgen, the biotechnology and pharmaceutical company.
Sharer, a 63-year-old former Navy engineer, was already earning lots of money — about $15 million in the previous year, plus such perks as two corporate jets.

The board decided to give Sharer more. It boosted his compensation to $21 million annually, a 37 percent increase, according to the company reports.

This outsize compensation has allowed Mr Sharer to live large.
He owns at least three homes, according to property records: a $2 million home in Los Angeles, a $6 million spread in Vail, Colo., and a $5 million place on Nantucket.
The Company's Less Than Stellar Performance

This huge raise on top of an already huge compensation package, however, seemed completely unrelated to the performance of the company Mr Sharer lead.
Shareholders at the company, one of the nation’s largest biotech firms, had lost 3 percent on their investment in 2010 and 7 percent over the past five years. The company had been forced to close or shrink plants, trimming the workforce from 20,100 to 17,400.

In addition,
by 2006, the growth that once distinguished Amgen had begun to taper off.

In 2008, Forbes magazine named ... [Sharer] among among the country’s most overpaid bosses because the company’s returns had dipped 4 percent annually over six years while he’d earned an average of $12.3 million.

The lack of enthusiasm was even more noticeable among the people who know most about Amgen — insiders. Companies are required to report when their officials sell stock. Since 2002, the ratio of share sales to purchases has been 12 to 1, according to Bloomberg data reviewed last fall by Christopher.

"Peer Benchmarking"

The Post article then analyzed the process used to set CEO pay at Amgen, and at many other corporations and organizations, and how it lead to such extreme results.
This is how it’s done in corporate America. At Amgen and at the vast majority of large U.S. companies, boards aim to pay their executives at levels equal to or above the median for executives at similar companies.

The idea behind setting executive pay this way, known as 'peer benchmarking,' is to keep talented bosses from leaving.

We have discussed examples of peer benchmarking before, most recently here. Yet I believe the Post article has the clearest explanation to date of how this works and what its results are. Note that:
It wasn’t until recently, however, that its pervasiveness and impact on executive pay became clear. Companies have long hid the way they set executive pay, but in late 2006, the Securities and Exchange Commission began compelling companies to disclose the specifics of how they use peer groups to determine executive pay.

Peer benchmarking has two components. First the pay of peers are assessed, that is, benchmarked:
Amgen selected 11 companies in the biotech/pharmaceutical field, which seems natural enough.

However, how the peers are chosen is important (just as how the comparator treatment is chosen in a clinical trial of a drug is important.) In this case,
But most of the companies on the list are far larger than Amgen. Amgen’s revenue in 2010 was $15 billion; the median revenue of its peer companies was $40 billion, according to Equilar.

The practice of choosing peers that boost pay is common. Studies by Faulkender, Bizjak and ISS Corporate Services have shown that when choosing 'peers' for pay-setting purposes, companies tend to choose larger firms or firms with more highly paid chief executives.

Then an assessment is made, almost always by the organizations board of directors or trustees, of how the CEO compares to these supposed peers:
Maybe even more significantly, however, the Amgen compensation committee also decided that Sharer, despite being at a smaller company, should earn stock compensation at the 75th percentile of peers.

This is critical because stock compensation tends to be the largest component of executive pay.

What is missing in the Post article is any rationale for the assessment that Sharer was doing so much better than the median CEO of these questionably comparable companies. The information presented in the article suggested instead that the company was doing badly, with decreasing revenue, decreasing share prices, and declining employment.

Cronyism on the Boards

Serving on the board of directors of a for-profit corporation requires board members to "demonstrate unyielding loyalty to the company's shareholders" [Per Monks RAG, Minow N. Corporate Governance, 3rd edition. Malden, MA: Blackwell Publishing, 2004. P.200.] (Boards of trustees of non-profit organizations have duties of loyalty, "a board member must give undivided allegiance when making decisions affecting the organization" and duty of obedience, "board members [must] ... be faithful to the organization's mission. They are not permitted to act in a way that is inconsistent with the central goals of the organization," per BoardSource.)

Rather than representing the interests of stockholders, however, many boards seem to put the interests of the CEO and top management first:
The practice has persisted because corporate board members, many of whom have personal or business relationships with the chief executive, have been unwilling to abandon the practice.

At Amgen, for example, four of the six members of the board compensation committee had personal or business connections to Sharer before joining the board. In fact, he nominated at least two of the six to the board, according to a company source and reports.

These kinds of ties — between chief executives and the boards that oversee them — permeate corporate America. On a typical board, the chief executive considers about about 33 percent of the board of directors as “friends” rather than as mere “acquaintances,”according to a survey of chief executives at about 350 S&P 1500 corporations conducted over 15 years by University of Michigan business professor James Westphal.

More tellingly, the chief executive is likely to find even more friends on the compensation committees of corporate boards — almost 50 percent.

The Obvious Problems with Peer Benchmarking

But the practice has long been controversial because, as critics have pointed out, if every company tries to keep up with or exceed the median pay for executives, executive compensation will spiral upward, regardless of performance. Few if any corporate boards consider their executive teams to be below average, so the result has become known as the 'Lake Wobegon' effect.

It wasn’t until recently, however, that its pervasiveness and impact on executive pay became clear. Companies have long hid the way they set executive pay, but in late 2006, the Securities and Exchange Commission began compelling companies to disclose the specifics of how they use peer groups to determine executive pay.

Since then, researchers have found that about 90 percent of major U.S. companies expressly set their executive pay targets at or above the median of their peer group. This creates just the kinds of circumstances that drive pay upward.

Moreover, the jump in pay because of peer benchmarking is significant. A chief executive’s pay is more influenced by what his or her 'peers' earn than by the company’s recent performance for shareholders, according to two independent research efforts based on the new disclosures. One was by Michael Faulkender at the University of Maryland and Jun Yang of Indiana University, and another was led by John Bizjak at Texas Christian University.

'Peer benchmarking has a significant influence on CEO pay,' Bizjak said. 'Basically, you can’t have every CEO paid above average without pay ratcheting upward over time.'

The gap between what workers and top executives make helps explain why income inequality in the United States is reaching levels unseen since the Great Depression.

Since the 1970s, median pay for executives at the nation’s largest companies has more than quadrupled, even after adjusting for inflation, according to researchers. Over the same period, pay for a typical non-supervisory worker has dropped more than 10 percent, according to Bureau of Labor statistics.

Critical to executive pay levels is peer benchmarking.

Even before the extent of the practice was known, it drew criticism from prominent business figures. After the Enron scandals, a blue-ribbon committee led by Peter G. Peterson, then chairman of the Federal Reserve Bank of New York, and John Snow, former chairman of the Business Roundtable, called for setting executive pay 'unconstrained by median compensation statistics.' Legendary investor Warren Buffett, in one of his famously plain-spoken letters to investors, likewise derided the method.

'Outlandish ‘goodies’ are showered upon CEOs simply because of a corporate version of the argument we all used when children: ‘But, Mom, all the other kids have one,’' he wrote.

Similarly, former Federal Reserve chairman Paul Volcker called it the 'Lake Wobegon syndrome' in congressional testimony in 2008, referring to Garrison Keillor’s fictional town where 'all the children are above average.'


Peer benchmarking, as it currently is implemented, is an irrational system that has nothing to do with the free market. It shows how perverse the incentives given to top health care leaders have become. It shows how top hired executives of health care organizations are no longer are accountable to anyone. The boards that are supposed to exercise stewardship over them have become their cheerleaders, possibly because they are populated by cronies, and current and former hired executives of other organizations who are more interested in preserving the wealth and power of hired executives than representing shareholders, or upholding organizations' missions.

The problems caused by such perverse incentives include:

- Paying the leaders so well (and supporting the public relations staff who sing their praises) costs money that could better be used to improve patients' and the public's health.

- Providing such perverse incentives, which seem entirely unrelated to the organizations' primary missions, is likely to distract leaders from the organizations' abilities to support these missions.

- It is likely that employees of such organizations who actually try to uphold the mission will be demoralized by the realization that their leaders' enrichment comes first, further damaging the organizations' ability to uphold the mission.

- Finally, it is possible that such perverse incentives attract leaders who are particularly unsuited to the task of upholding the mission.

So, ad infinitum, I repeat.... health care organizations need leaders that uphold the core values of health care, and focus on and are accountable for the mission, not on secondary responsibilities that conflict with these values and their mission, and not on self-enrichment. Leaders ought to be rewarded reasonably, but not lavishly, for doing what ultimately improves patient care, or when applicable, good education and good research. On the other hand, those who authorize, direct and implement bad behavior ought to suffer negative consequences sufficient to deter future bad behavior.

If we do not fix the severe problems affecting the leadership and governance of health care, and do not increase accountability, integrity and transparency of health care leadership and governance, we will be as much to blame as the leaders when the system collapses.