The Massachusetts Attorney General’s office issued a report on its investigation of Blue Cross Blue Shield’s termination of former CEO Cleve Killingsworth and paying him $4,260,000 in severance. Original press reports said that Killingsworth was paid $11 million in severance, but much of that was for a retirement benefit earned over prior years. It’s no wonder that this caught the attention of the press or that the Attorney General decided to investigate. What is surprising, though, is that the AGO’s report criticizes use of executive employment agreements and paying severance when firing an executive for less than fully satisfactory performance. One gets the impression that the folks in the AG’s office have never looked at CEO contracts and didn’t know that they all promise severance on involuntary termination without cause, that most employment agreements use automatic renewals, and that almost all of them call for severance if the contract is not renewed. These are the facts. Killingsworth was hired by BCBS in February 2004 as its COO, and his offer letter promised him severance if he were terminated without cause. A year later he was promoted to CEO and given a contract promising severance if he were involuntarily terminated without cause. Severance was set at two years’ salary, target incentive award, and health and life insurance, as well as outplacement and financial planning services. His target annual incentive award was equal to his base salary, so two years’ pay was four times his annual salary. His salary was $1.04 million, so the total was over $4 million. Killingsworth resigned after a period of weak financial performance, at the request of the board, but for reasons that did not amount to “cause” as defined in his contract (dishonest acts, willful violation of regulations or policies or code of conduct, commission of or conviction of a felony or a misdemeanor involving moral turpitude, deceit, dishonesty, or fraud; or a material breach of obligations under the contract). While CEO, Killingsworth served on boards of as many as 14 other organizations, and earned significant compensation from three of them. Under criticism and pressure from the AGO, BCBS decided to rebate $4.6 million to policy holders. After first concluding that BCBS owed him this severance, the AGO’s report goes on to criticize the settlement. Even on termination for poor performance, the board was still obligated to pay severance. The automatic renewals entitled Killingsworth to remain in the job as long as he wanted, or until death, disability, or voluntary retirement, unless the board was willing to pay severance. Paying severance could be avoided only if the reasons for involuntary termination involved intentional misconduct. In the course of the investigation, the AGO obtained and reviewed CEO contracts from other local health care providers and insurers. It found that all agreements promised severance, and that they all limited the circumstances under which the board could terminate the CEO without paying severance to acts involving intentional misconduct. After admitting that severance might be important to protect a new executive, the report suggests that severance serves no important purpose after the executive has been in the job for a while, that severance shouldn’t be necessary for someone promoted internally or someone who doesn’t need to relocate to take the job, and that severance should be paid on termination for poor performance. It goes so far as to say that “the AGO doubts any [executives of our large hospital systems and health systems] were such reluctant warriors as to require these protections before accepting the position.” It also concludes that boards’ concern over the cost of severance or public reaction to it could lead a board to delay termination of an executive, which impedes the board’s ability to protect the organization. Having found something it doesn’t like, the AGO decided that it will require tax-exempt health care providers and insurers to disclose all employment agreements with senior executives, with an explanation of the basis for the protection provided. Totally aside from the purpose of the investigation, the AGO also decided to establish a new standard for nonprofits for CEO performance appraisal and for CEO service on other boards, if the boards pay their board members. The report asserts that no board is really competent to perform an adequate performance review without external professional assistance; and that a CEO’s performance appraisal should include evaluation by subordinates and outside contacts. It implies that a CEO’s service on a board, if board members are paid, as they are on for-profit boards, amounts to selling a charitable asset, which is implicitly a form of private inurement. The prestige associated with the CEO’s position at a charitable organization is an asset of the charitable organization; and the CEO’s invitation to be a paid member of a for-profit board comes from the prestige of the position, not the prestige of the individual, so it amounts to selling a charitable asset for personal gain. Wow! This reminds me of a thunderous sermon from an uptight, pious Puritan preacher. Yes, the $4 million in severance probably was too much. Yes, BCBS shouldn’t pay severance on voluntary resignation—but it wasn’t really voluntary. The practice of allowing executives to resign to save face yet still collect severance makes perfect sense in some ways, but it violates the terms of the contract and amounts to making a parting gift to a fired CEO (which would be an excess benefit transaction subject to intermediate sanctions, by the way, if it occurred at a 501(c)(3) charity). But boards agree to pay severance for several perfectly good reasons, aside from the fact that they generally can’t recruit outsiders as CEOs without a contract offering severance on involuntary termination without cause (with cause defined narrowly to exclude poor performance). One is that it actually makes it easier to terminate an executive, rather than harder, as the AGO assumes, since there’s no reason to feel bad about firing someone and no reason to worry about legal claims if you pay enough severance to buy a release from claims of discriminatory termination and enough to cover whatever guilt or embarrassment may arise in terminating a long-service CEO who has generally performed well. (Killingsworth was, after all, a member of two protected classes and a colleague of the board members, as well as former chair of the board.) Severance is both a form of liquidated damages for breaking a contract and a form of deferred compensation intended to provide income continuity in case of getting fired for no particularly good reason. Many reasons for terminating executives have to do with issues that are a matter of perception, after all, and much of what is perceived as a performance problem is due to externalities. It’s true that employment agreements aren’t needed for CEOs any more than they are for the multitude of employees who toil without contracts. But it helps to have an agreement about what the penalty will be for firing a long-term employee (who must have been performing reasonably well, to survive in the job for a long time) for no good reason other than the board deciding that it is time to ask the CEO to leave. Employment agreements also help retain executives, who can easily move to other jobs, almost whenever they want, and usually for equal or higher pay. The employment agreement is a request from the board that the CEO promise to stay for a number of years, at a time when the CEO could just as well accept other job offers. If the board then changes its mind, it is breaking a contract, and severance is the proper penalty for breaking the contract. The board has enticed the CEO to stay with the organization and turn down other opportunities to move, with a promise of employment for a period of years. By the time the board changes its mind, the CEO has already turned down other attractive opportunities that are no longer available. Absent employment agreements and promises of severance, executives would move around more often. They help employers retain good executives, every bit as much as they make it easy to terminate executives when it seems time to do so. Four times salary may be excessive severance, because the underlying justification for a particular amount of severance is to keep income more or less whole until the executive finds comparable employment. It might take a CEO a year or more to find a truly comparable position in a place he wants to work, but it shouldn’t take four years. Severance is often based on total cash compensation (salary plus bonus), but that is hard to stomach when the salary is $1 million and the bonus is another $1 million. Insofar as the purpose of severance is income continuity, continuing salary for a year or two should be sufficient. Insofar as the purpose is to settle damages for breaking a contract (and for having given up opportunities to take other positions elsewhere), the damages probably should be some multiple of total compensation, particularly if the termination damages the executive’s reputation and makes it harder to find the next job, or if the termination occurs so close to normal retirement age that it more or less amounts to forced early retirement. Don’t forget that Congress set the bar for severance when it declared that severance is deductible only up to 2.99 times total compensation. That, in effect, blessed unnecessarily generous severance, and it led to a rapid increase in the amount of severance offered to executives. Over time, boards have become more conservative about severance, and two times’ total pay is now more or less the norm for CEOs of large organizations. How, though, the AGO thinks it is an expert on CEO performance appraisal, is another matter. I doubt that any of the attorneys who wrote the report know much about the range of processes used for performance appraisal. They obviously don’t see the inconsistency between saying that trustees aren’t competent to appraise the CEO’s performance without external assistance and the premise that they are capable of governing the organization. Most of the BDBS-Massachusetts board members are senior executives within their own businesses and undoubtedly have far more experience with performance appraisal than the attorneys in the AG’s office. They presumably know what they want from a leader and use that in evaluating the CEO’s performance. They may not be omniscient, but no external facilitator is going to be, either.