Please read the article attached and answer these questions
98 Harvard Business ReviewMay–June 2021
Claudio
Fernández-
Aráoz
Executive fellow
for executive
education, Harvard
Business School
Gregory
Nagel
Associate
professor of
finance, Middle
Tennessee
State University
Carrie Green
Director of
equities,
Tennessee
Consolidated
Retirement
System
PHOTOGRAPHER BALINT ALOVITS
AUTHORS
L E A D E R S H I P
D E V E L O P M E N T
Harvard Business Review
May–June 2021 99
The High
Cost of Poor
Succession
Planning
A better way
to find your
next CEO
L E A D E R S H I P
D E V E L O P M E N T
At the time Microsoft was the third-most-profitable
company in the United States and the fourth most valuable.
Nevertheless, this well-respected global technology giant
didn’t seem to have a plan for replacing Ballmer, even though
he had, according to most informed observers, underper-
formed for years. (Critics cite his slow move into mobile,
social media, and video along with ill-fated acquisitions
and product reboots.) While a few high-profile executives,
such as Windows chief Steven Sinofsky and Xbox head Don
Mattrick, had jumped ship during his tenure—another sign
of trouble—with a workforce of 100,000, Microsoft surely
could have identified other promising candidates in senior
management roles, not to mention outsiders, who’d be ready
to step in for Ballmer.
Instead, Microsoft seemed to start from square one,
concentrating mostly on external candidates. According to
the director who chaired the search committee, the board
cast a wide net across a number of industries and skill sets,
identified more than 100 candidates, talked with several
dozen, and then focused intensely on about 20. Among
them was Steve Mollenkopf, the COO of Qualcomm, who
fell out of contention when he was promoted to that compa-
ny’s top job. Alan Mulally, who had just turned around Ford
and was the favorite candidate, took his name off the list
in January—at which point the press described Microsoft’s
board as turning to Plan B.
Finally, in February, six months after Ballmer had declared
himself a lame duck, Microsoft announced that an insider,
Satya Nadella, would become the third CEO in its history.
We know now that despite that bumbling succession
process, Nadella was a terrific pick. He moved Microsoft
away from fiefdoms and a “know-it-all” culture and toward
a more open, collaborative “learn-it-all” one; built up the
cloud-computing business; made Office available on all
smartphones; and executed dozens of accretive acquisitions,
including the purchase of LinkedIn. In his first nine months
as CEO, Microsoft’s stock rose 30%, increasing its market
value by $90 billion. As we write this, seven years into his
tenure, it is the world’s second-most-valuable company.
But what if Microsoft hadn’t promoted Nadella? What
if its hastily put together, extremely broad, and externally
focused search had resulted in the hiring of an outsider?
What if Mulally, who had no tech sector experience, had
been appointed? Why hadn’t the board already been
grooming Nadella—a 21-year veteran of the company with
clear leadership competence, cultural fit, and expertise in
In August 2013, Steve Ballmer abruptly announced that he
would step down as chief executive of Microsoft as soon as
his replacement could be found. Thus began one of the most
important CEO searches in the past decade—and a case
study in the dos and don’ts of senior leadership succession.
100 Harvard Business ReviewMay–June 2021100 Harvard Business ReviewMay–June 2021
THE ADVICE
Companies—and especially their
directors—must plan leadership changes
before they’re needed, identify and
develop rising stars, give them access
to the board, look at both internal
and external candidates, and partner
cautiously with executive search firms.
I D E A I N B R I E F
THE PROBLEM
Many large companies fail to pay
adequate attention to their top-
level leadership pipelines and suc-
cession processes, which results in
excessive turnover and significant
value destruction for companies
and investment portfolios.
THE RESEARCH
Analysis suggests that the market value wiped
out by badly managed CEO and C-suite
transitions in the S&P 1500 alone is close to
$1 trillion a year. Better succession planning
could, by contrast, help the large-cap U.S.
equity market add a full point to the 4% to 5%
annual gains Wall Street projects for it.
ABOUT THE ART
Balint Alovits’s project Time Machine explores Budapest’s
Bauhaus and art deco spiral staircases, using perspective
and the repetition of form to evoke a sense of infinity.
Harvard Business Review
May–June 2021 101
up-and-coming areas of technology—or any of his similarly
qualified peers?
While Microsoft did make the right decision in the end, its
lack of planning could have led to a costly disaster.
Like Microsoft, many large companies fail to pay ade-
quate attention to their leadership pipelines and succession
processes. And most of them don’t get as lucky as Microsoft
did. In our combined nine decades of experience in execu-
tive search and talent development (Claudio), professional
investment (Carrie), and management and financial research
(Gregory), we’ve seen flawed succession practices lead to
excessive turnover among senior executives and, in the end,
significant value destruction for companies and investment
portfolios.
In our recent research we’ve attempted to quantify those
costs. According to our analysis, the amount of market value
wiped out by badly managed CEO and C-suite transitions in
the S&P 1500 is close to $1 trillion a year. We estimate that
better succession planning could help the large-cap U.S.
equity market add a full point to the 4% to 5% annual gains
that Wall Street projects for it. In other words, company
valuations and investor returns would be 20% to 25% higher.
In this article we’ll examine those findings and then make
recommendations for how to significantly improve corporate
performance and investor returns through better practices
for grooming and selecting CEOs. Of course, these lessons can
apply to succession planning for other key senior manage-
ment roles as well.
QUANTIFYING THE PROBLEM
In our opinion large companies’ excessive tendency to hire
leaders from outside is one of the biggest problems with
succession practices. This propensity incurs three major
kinds of costs: underperformance at companies that hire
ill-suited external CEOs, the loss of intellectual capital in the
C-suites of the organizations that executives leave behind,
and for those companies promoting from within, the lower
performance of ill-prepared successors.
A landmark study that Rakesh Khurana and Nitin Nohria
of Harvard Business School conducted years ago sheds light
on the first kind of cost. Khurana and Nohria examined
the impact that different types of CEO succession had on
operating returns in 200 organizations over a 15-year period.
They compared four scenarios: (1) an insider promoted in a
firm doing reasonably well; (2) an insider promoted in a firm
doing poorly; (3) an outsider hired in a firm doing reasonably
well; and (4) an outsider hired in a firm doing poorly. They
found that, on average, insiders didn’t significantly change
their company’s performance. That makes sense: Similar
people working in similar ways at the same company will
produce similar results. With outsiders, the change was much
more extreme. In the infrequent cases when a company was
doing very poorly, outsiders added great value, on average.
But at companies doing reasonably well, outsiders destroyed
massive value. This suggested that companies looking for a
new CEO should hire external candidates only in exceptional
cases, when a major turnaround or cultural change is called for.
Other research has confirmed that external hiring usually
doesn’t deliver on its promise. For example, Matthew Bidwell
of the Wharton School of Business found that while outsiders
often appear to have better experience and education than
insiders do, they are paid more, perform worse, and have
higher exit rates. Additional studies support that: One by
Cláudia Custódio, Miguel Ferreira, and Pedro Matos showed
that external CEO hires were paid 15% more than internal
hires, on average; and one by Sam Allgood and Kathleen
Farrell revealed that CEOs brought in from the outside have
an 84% greater chance of turnover than insiders in the first
three years, usually for poor performance.
Another recent study found that companies often choose
outsiders because they have already served as CEOs else-
where—indicating the firms value previous experience in the
role over insiders’ potential to excel. But that experience rarely
guaranteed success: When the researchers looked at S&P 500
CEOs who had led more than one company, they found that
70% had generated better performance the first time around.
Despite those downsides, S&P 1500 companies hired
their CEOs from outside 26% of the time from 2014 to 2018,
according to ExecuComp data—perhaps because, as Whar-
ton’s Peter Cappelli has found, companies have an irrational
bias toward exciting and unblemished external hires whom
they know less about.
We wanted to investigate how external CEOs performed
relative to what insiders might have done in the same
positions. Without the ability to rewind time and play out
L E A D E R S H I P
D E V E L O P M E N T
102 Harvard Business ReviewMay–June 2021
different scenarios, that would seem impossible to do.
However, we believe that with statistics, we can predict what
would have happened with different CEO hires.
We used a technique known as structural self-selection
modeling (SSSM), directly derived from Nobel Prize winner
James Heckman’s research. It is similar to the multiple
regression modeling that companies frequently employ in
forecasting and scenario-planning exercises. We first iden-
tified 80 independent variables, including firm characteris-
tics (like size and capital expenditures), sector, risk, board
structure, and short- and long-term performance before and
after a change in CEOs. The performance metric we used was
cash-flow return on assets, which unlike operating return on
assets accounts for the reorg and restructuring costs that are
frequent following the arrival of an outsider CEO.
We then looked at every instance in which an outsider
CEO was hired to lead a public U.S. firm over a 17-year period
and calculated the change in cash-flow return on assets for
his or her tenure. We plugged the 80 independent variables
for each of those companies into the SSSM to create a “coun-
terfactual”: what the expected change in cash-flow return
on assets would have been if the company had promoted an
insider. We found that only 39% of outside hires would have
done better than a theoretical inside hire.
Of course, nobody knows in advance what the perfor-
mance of any appointed executive will be. But boards should
base consequential and risky hiring decisions on their best
estimate of future outcomes. Our analysis shows that in only
7.2% of instances will an outside CEO hire have a 60% chance
of outperforming an insider, and in a mere 2.8% of cases will
he or she have a 90% chance of outperforming an insider.
Dramatic as those figures are, they tell only part of the
story. One key knock-on effect of external choices for CEO and
other senior positions is the loss of intellectual capital in the
C-suites of the firms those executives were hired from. And
because on average executives perform worse at the company
they jump to, the negative impact on the entire market is
even greater. We can calculate the effect that loss of intellec-
tual capital has on market valuations by both analyzing the
impact of sudden CEO departures and using the economic
model provided by Hanno Lustig, Chad Syverson, and Stijn
Van Nieuwerburgh to track how much intellectual capital a
departing manager can transfer to his or her next employer.
Our analysis shows that the decrease in intellectual
capital at new executives’ previous employers leads to a
0.7 percentage point reduction in total shareholder returns
for the S&P 1500, or $255 billion, each year. When we add
in the underperformance at the firms hiring external CEOs,
total shareholder returns fall by about another half a per-
centage point, costing investors an additional $182 billion.
The final impact, where companies do promote CEOs from
within but fail to properly prepare them to take over, costs
an additional 0.3 percentage point, bringing the total loss
across the S&P 1500 portfolio to $546 billion. To calculate
the third cost, we drew from a study of 2,900 companies
done by Olubunmi Faleye of Northeastern University, which
found that the return on assets of firms with poorly prepared
internal CEO successors is significantly lower than that of
firms that properly prepared them. A simple extrapolation
of these findings to global equity markets, collectively worth
about $58 trillion at the time of this writing, implies that the
total annual costs to global shareholders would amount to
$870 billion. This global estimate is probably conservative,
given that governance, succession, and talent practices
usually are significantly better in the United States than in
most other countries. We’re currently extending our analysis
to other major equity markets to try to confirm it.
Another negative by-product of poor succession planning
and excessive outside hiring is rising CEO compensation as
companies compete for the same top executives. Financier
Worldwide reported that at the top 350 U.S. companies, aver-
age CEO pay had climbed to $17 million in 2018, or about 278
times a typical employee’s compensation. From 1978 to 2018,
CEO pay had jumped by more than 1,000%, while the average
worker’s pay had risen just 12%. Though those figures are
shocking, our analysis shows that skyrocketing CEO compen-
sation actually plays only a small role in value destruction.
The main costs of ill-considered successions remain poor
performance by outsider CEOs, loss of C-suite intellectual
capital at the firms that CEOs and other top executives leave
behind, and ill-prepared internally promoted executives.
One final note: We intentionally focused this analysis on
large firms because we believe that’s where the problem of
poor succession at the top is most acute. Small firms usually
lack a deep talent pool, so they can be better served by hiring
CEOs from the outside.
Large companies’ excessive tendency to hire leaders from outside is
one of the biggest problems with succession practices.
Harvard Business Review
May–June 2021 103
IMPLEMENTING SOLUTIONS
Why are some of the world’s biggest and most powerful
organizations getting CEO appointments so wrong? For five
main reasons: lack of attention to succession, poor leadership
development, suboptimal board composition, lazy hiring
practices, and conflicted search firms. Here are some recom-
mendations for fixing those problems.
Plan succession well before you think you need to.
According to PwC’s latest Strategy& “CEO Success” study,
in 2018 turnover among CEOs at the world’s largest 2,500
companies reached nearly 18%—the highest rate PwC had
ever tallied. A disturbing 20% of those departing CEOs were
forced out, and for the first time in the study’s history, more
CEOs were dismissed for ethical lapses than for financial
performance or conflicts with their boards. Looking forward,
we suspect that unanticipated CEO turnover will continue to
rise because of the growing attention to moral issues (such as
sexual harassment) and industry and market volatility.
Despite this trend, boards continue to be caught off guard
because they haven’t spent enough time developing talent
and mapping out possible lines of succession. Some believe
that having a casual “if the CEO gets hit by a bus tomorrow”
plan, which picks a replacement but doesn’t prepare or vet
that person or weigh alternatives, is enough. It is not. Others
delegate succession planning to the CEO, which is an equally
unacceptable abnegation of duty. For instance, we know of a
major company, valued at hundreds of billions of dollars, with
a CEO in his late sixties who has been unwilling to properly
develop any potential replacements. Unfortunately, because
the firm’s recent results and stock market performance have
been good, board members are afraid to confront him.
Succession planning should start the moment a new CEO
is appointed. Take Ajay Banga, the former chief executive
and current chairman of Mastercard: He began discussing
when he might cede the CEO role to a successor even as he
was interviewing for the job himself. The process should
remain robust, with directors constantly monitoring and if
need be adjusting the pipeline. If there isn’t already a poten-
tial successor among the CEO’s direct reports, the board
should look to the next level and consider advancement and
development opportunities that will help executives there
progress. If that level is empty, directors can promote or hire
high potentials into it or the C-suite. While hiring externally
L E A D E R S H I P
D E V E L O P M E N T
104 Harvard Business ReviewMay–June 2021
Harvard Business Review
May–June 2021 105
is usually not ideal, it’s much less risky to do it at a lower
level than in the top job.
Purposefully identify and develop your rising stars.
By now most directors know the attributes and skills that
senior executives need. At the leadership advisory firm Egon
Zehnder, where one of us (Claudio) worked for three decades,
the list used for CEO searches includes intelligence and
values. The firm also assesses candidates on strategic orien-
tation, market insight, results focus, and customer impact,
and their competence at collaborating with and influencing
others, organizational development, leading teams, and
change management. Meaningful succession planning calls
for finding rising managers who either have the right levels of
all those capabilities or, more likely, the potential to develop
them. Four critical traits—curiosity, insight, engagement,
and determination—signal potential, and with the proper
coaching and support, people who demonstrate them can be
groomed for high-level positions. (For more on this subject,
see “Turning Potential into Success: The Missing Link in Lead-
ership Development,” HBR, November–December 2017.)
One important development area for any CEO is emotional
intelligence, which encompasses flexibility, adaptability,
self-control, and relationship management. You might think
that those soft skills would be more challenging to learn than
hard ones such as calculus or coding. But as Richard Boyatzis
of the Weatherhead School of Management has conclusively
demonstrated, people can pick up these crucial leadership
competencies even as adults.
Another way for boards to help potential successors get
ready is to insist that they be given challenging rotations and
stretch assignments, as was common at General Electric in its
glory days and is practiced with great success at Unilever and
McKinsey today. When you expose your highest potentials to
new geographies, businesses, situations, and functions, you
can become a leadership factory.
Appoint the most promising executives to the board—
or give them more access to it. In the United States, in
part because of regulatory mandates following executive
malfeasance at Enron, Tyco, and other companies, most large
companies’ boards have become fully independent, with the
CEO as the only employee director. Faleye found that the pro-
portion of U.S. boards set up this way exploded from about
a third in 1998 to more than two-thirds in 2011. Our analysis
shows that the percentage of fully independent boards has
continued to increase, rising to 76% by 2018.
While there are clear benefits to getting oversight and
advice from outside experts, we believe independent boards
are less equipped to manage CEO succession. With so little
exposure to internal up-and-comers but extensive knowledge
of potential external hires from their own organizations and
other board experiences, directors are understandably more
likely to favor outside CEO candidates or be unduly influenced
by individual opinions. As one veteran director recently told
us, “It’s scary to see how little insight boards have about top
internal executives these days; a lot of the views are painted,
either too positively or too negatively, by the sitting CEO.”
We believe that boards should make room for one to three
executives who are potential successors to the CEO. Not only
does that allow directors to see likely candidates in action,
but it better prepares those individuals to take on the top job.
When Faleye compared the performance of internally pro-
moted CEOs who had prior director experience against that
of insiders who lacked it, he saw that during their first two
years the CEOs with board experience had an average return
on assets that was 12.5 percentage points higher. Interestingly,
this massive difference disappeared during year three, sug-
gesting that while both types of executives had similar levels
of competence and potential, the exposure to strategic board-
level discussions as well as the relationships established with
directors drastically flattened learning curves.
Indra Nooyi, for example, joined PepsiCo’s board when
she was the company’s CFO—five years before becoming its
CEO. Watching her firsthand, the board became confident in
her competence and potential and, after her appointment as
CEO, was more open to her plans to radically transform the
company by expanding its portfolio beyond sugary drinks and
steering it toward greater social responsibility. During Nooyi’s
tenure as CEO, PepsiCo’s net profit increased 122%.
If you have too many directors already or too many
promising potential CEO successors in your ranks, an alter-
native (though suboptimal) approach is to ask your rising
stars to frequently attend and present at board meetings.
This will improve their exposure, contributions, and devel-
opment. Before the pandemic, good boards ran dedicated
off-sites or group trips where directors and top executives,
and even their spouses, could connect professionally and
L E A D E R S H I P
D E V E L O P M E N T
106 Harvard Business ReviewMay–June 2021
personally. As boards get back into their rhythms post-
Covid, we hope that such in-person social interaction will
resume. For further development, you might also encour-
age some of your most likely successors to selectively join
other companies’ boards.
Look at internal and external candidates. The best
practice is to carefully outline your ideal CEO profile and then
look both inside and outside for the person who best matches
that description. While we believe that every company
should first master the art of spotting internal talent and
create succession plans based on its current roster, we also
see value in external searches for benchmarking and compre-
hensiveness. (And so do companies like Mastercard, PepsiCo,
P&G, and American Express.) Research from the Center
for Creative Leadership has consistently shown that when
companies consider wide pools of insiders and outsiders,
executive appointments are more successful. Whether you’re
shopping for a house or for your next top executive, compara-
tive evaluations produce better decisions.
Make sure to conduct thorough assessments of all can-
didates, even the insiders who are well known to the board.
Consider not who has performed the best until now but who
is ready to meet the future challenges of the CEO role and has
the potential to continue adapting in a volatile, uncertain,
chaotic, and ambiguous world. Judge everyone against your
job specs, grill candidates in well-structured interviews, and
conduct in-depth reference checks. This is the only way to
avoid appointing the wrong people to the job.
If you partner with search consultants, avoid the usual
perverse incentives. Executive search firms can usually add
great value to succession efforts. Consultants with the right
training and experience can identify the competencies that
each senior position requires, get more out of interviews and
reference checks, and distinguish potentially great perform-
ers from the rest. Such consultants also tend to have trusting
relationships with candidates, sources, and references.
However, the search profession as a whole still probably
hurts as much as it helps, owing to two blatantly perverse
incentives: the contingency arrangement and the percentage
fee. Most search consultants are compensated when they
produce a hire, regardless of that person’s fitness for the
job or origin. They make no money on inside hires, who
don’t need to be found and brought in. Traditionally, search
consultants are paid a third of the new executive’s annual
cash compensation (salary plus bonus). As a result, whether
consciously or unconsciously, many oversell high-priced
outsiders and shoot down internal alternatives. The solu-
tion is to swap the percentage fee with a prearranged fixed
fee that’s based on the importance of the position and the
complexity of the search and to replace the contingency fee
with a retainer so that the consultant is paid the same no
matter who is appointed. (Of course, the retainer fee makes
financial sense only if you’re planning to use the consultant
for enough search and advisory work to justify the cost.)
Even if you have those two things right, you should
still use search consultants only in special situations—for
example, if your internal candidates are unsuitable, you can’t
identify or access appropriate external candidates on your
own, or your company is entering a new business, region, or
period of strategic change. Then approach the selection of
your consultant as you would any other people decision: Ask
for recommendations, consider multiple firms, and check
references. Once you’ve developed a short list, meet the
recruiters in person to get a read on their relevant experience,
as well as their level of professionalism, candor, and concern.
C O M P A N I E S A N D I N S T I T U T I O N S must do a better job of
getting CEO succession right—their organizations, their
industries, and their market returns depend on it. We hope
this article helps senior executives, directors, and investors
recognize the magnitude of the problem and act accord-
ingly. Microsoft shouldn’t have required a long and public
search to conclude that Nadella was the right leader to get
the company back on track after Ballmer’s years of struggle.
It should have already had him—and even other potential
successors—waiting in the wings. How many rising stars like
Nadella do you have at your company—and what can you do
tomorrow to put them on a path to becoming your next (and
ideally best ever) CEO? HBR Reprint R2103F
CLAUDIO FERNÁNDEZ-ARÁOZ is an executive fellow for execu –
tive education at Harvard Business School and the author of
It’s Not the How or the What but the Who (Harvard Business Review
Press, 2014). GREGORY NAGEL is an associate professor of finance
at Middle Tennessee State University. CARRIE GREEN is the director
of equities for the Tennessee Consolidated Retirement System.
Boards should make room for one to three executives who are potential successors to
the current CEO. Board experience helps prepare those individuals to take on the top job.
Harvard Business Review
May–June 2021 107
Copyright 2021 Harvard Business Publishing. All Rights Reserved. Additional restrictions
may apply including the use of this content as assigned course material. Please consult your
institution’s librarian about any restrictions that might apply under the license with your
institution. For more information and teaching resources from Harvard Business Publishing
including Harvard Business School Cases, eLearning products, and business simulations
please visit hbsp.harvard.edu.
