Chapter 12 Agency Problems, Compensation, and Performance Measurement 311
bre44380_ch12_302-326.indd 311 09/11/15 07:55 AM
not managers or employees and not linked to the company by some other relationship—for
example, a lucrative consulting contract—that would undercut their independence. The com-
mittee typically hires outside consultants to advise on compensation trends and on compensa-
tion levels in peer companies.
You can see how compensation tends to creep up. The problem is that boards don’t want to
approve below-average compensation. But if every firm wants to be above-average, then the
average will ratchet up.^11
Once the compensation package is approved by the committee, it is described in an annual
Compensation Discussion and Analysis (CD&A), which is sent to shareholders along with
director nominations and the company’s 10-K filing. (The 10-K is the annual report to the
SEC.) On January 2011, the SEC gave shareholders a nonbinding yes-or-no vote on the CD&A
at least once every three years.^12 The occasional no vote on management compensation is a
disagreeable wake-up call for managers and directors. For example, when the shareholders of
Charles River Laboratories voted no in 2013, the company made a number of changes to the
compensation package before seeking (and obtaining) shareholder approval in 2014.
Yet these safeguards don’t satisfy everyone. Just as Dr. Seuss predicted, we now have pay-
watcher-watchers, such as the consulting company ISS. ISS reviews CD&As for thousands
of companies, looking especially at pay-for-performance standards. ISS’s clients are mostly
institutional investors, who seek advice on how to vote. (A mutual fund or pension fund may
own shares in hundreds of companies. The company may decide to outsource the analysis of
CD&As to a specialist company like ISS.)
12-2 Measuring and Rewarding Performance: Residual Income and EVA
Almost all top executives of firms with publicly traded shares have compensation pack-
ages that depend in part on their firms’ stock price performance. But their compensation
also includes a bonus that depends on increases in earnings or on other accounting measures
of performance. For lower-level managers, compensation packages usually depend more on
accounting measures and less on stock returns.
Accounting measures of performance have two advantages:
- They are based on absolute performance, rather than on performance relative to inves-
tors’ expectations. - They make it possible to measure the performance of junior managers whose responsi-
bility extends to only a single division or plant.
Tying compensation to accounting profits also creates some obvious problems. For exam-
ple, managers whose pay or promotion depends on short-term profits may cut back on train-
ing, advertising, or R&D. This is not a recipe for adding value because these outlays are
investments that should pay off in later years. Nevertheless the outlays are treated as current
expenses and deducted from current income. Thus an ambitious manager is tempted to cut
back, thereby increasing current income, leaving longer-run problems to his or her successor.
In addition, accounting earnings and rates of return can be severely biased measures of true
profitability. We ignore this problem for now, but return to it in the next section.
(^11) Bizjak, Lemmon, and Naveen found that most firms set pay levels at or above the median of the peer group, and some firms go
much higher. For example, Coca-Cola and IBM consistently aim for levels in the top quartile of their peers. See J. M. Bizjak, M. L.
Lemmon, and L. Naveen, “Has the Use of Peer Groups Contributed to Higher Pay and Less Efficient Compensation?” Journal of
Financial Economics 90 (November 2008), pp. 152–168.
(^12) Other countries that have given shareholders nonbinding votes on compensation include Australia, Sweden, and the U.K. Sharehold-
ers in the Netherlands have a binding vote.