CEO Succession — Insiders, Outsiders, Mandates

Every module so far has asked what kind of CEO a company needs. This one asks the question at the moment it is actually decided: succession. It is the highest-leverage fit decision a board makes, and most boards make it badly for a predictable reason — they argue about insider versus outsider before they have written down the problem.

The research is clear on one thing and honest about the rest. Outsiders generate more strategic change than insiders. Whether more change is good depends on how much the company needed, how much it can absorb, and what happens to the senior team meanwhile. There is no general insider or outsider advantage; there are conditions under which each pays.

Three variables do the work. Initial disruption: an outsider arrives without the relationships and tacit knowledge that make an organization run. Institutional knowledge: an insider has it — an asset when the problem is execution, a liability when the problem is the institution itself. Mandate: what the board actually wants changed, which is often unstated and sometimes unknown to the board. Family succession gets its own treatment because the evidence is unusually strong and unusually uncomfortable.

In the Effectiveness Equation — Traits × Behaviors × Organizational Context × Current Moment — succession is where the Current Moment term is set for the next several years. The board is not choosing a person. It is choosing which moment the company is in.

Outsiders change more; more change is not better. The right successor is the one whose mandate matches the problem and whose integration costs the company can afford.

Strategic change has an inverted-U relationship with performance, and outsiders live at both ends of the curve — larger gains from moderate change, larger losses from too much. The succession decision is therefore not "who is best?" but "what dose of change does this company need, who can deliver that dose, and what will it cost to get them productive?"

RESEARCH FINDING Karaevli (2007) reconceptualized insider/outsider status as a continuum — "outsiderness," reflecting distance from the firm and the industry — and tested it on longitudinal archival data from the U.S. airline and chemical industries, 1972–2002. There was no direct main effect of outsiderness on post-succession performance. Outsiderness helped when pre-succession performance was poor and when the environment was turbulent, and its effect depended on post-succession context, including concurrent strategic change and top-team turnover. Two industries, archival proxies, observational design. There is no general outsider advantage; outsiders help mainly when the firm is in trouble or the environment is changing, and the effect depends on what else changes.

RESEARCH FINDING Zhang & Rajagopalan (2010), studying 193 U.S. CEOs who departed between 1993 and 1998, found that the level of strategic change (shifts in resource-allocation patterns) has an inverted-U relationship with firm performance — moderate change helps, excessive change hurts. Both the upside of moderate change and the downside of excessive change were more pronounced for outside CEOs, and supplementary analysis showed the pattern is driven by later tenure years rather than the initial post-appointment period. Modest sample, accounting-based performance, change proxied by resource allocation. Outsiders amplify the curve in both directions, and the damage from overreach arrives later than boards expect.

RESEARCH FINDING Shen & Cannella (2002), drawing on power-circulation theory and 228 U.S. successions, distinguished "followers" (insiders succeeding a CEO who left voluntarily), "contenders" (insiders succeeding a dismissed CEO) and outsiders. Successor type interacted with post-succession senior-executive turnover to affect ROA, and departing-CEO tenure had an inverted-U relationship with post-succession ROA. Archival, modest sample. A succession is a team event, not an individual one.

RESEARCH FINDING Georgakakis & Ruigrok (2017), using 109 CEO succession events at large international firms, found outsider succession associated with better post-succession performance under specific conditions: when the outsider is demographically similar to the incumbent top team (easing integration), when the outsider has a diverse career background, and when the firm is already performing well in a favorable industry environment. INTERPRETATION Read alongside Karaevli, this looks contradictory — outsiders help in trouble; outsiders help when things are good — until you notice that both are about integration cost. A troubled firm has less to lose from disruption; a healthy firm can afford to pay for it. The firm that cannot afford an outsider is the one in the middle, doing fine, with a team that would resist.

RESEARCH FINDING Family succession has the strongest evidence in the module. Pérez-González (2006), in roughly 300-plus CEO successions at U.S. family-controlled public firms, found that firms appointing a family-related successor experienced large declines in return on assets and market-to-book relative to firms promoting unrelated CEOs, with underperformance concentrated among heirs who did not attend a selective undergraduate institution — consistent with nepotism limiting the talent pool. The selective-college proxy is coarse and succession choice is not random.

RESEARCH FINDING Bennedsen, Nielsen, Pérez-González & Wolfenzon (2007) addressed the selection problem with an instrumental-variables design on Danish administrative data: the gender of the departing CEO's first-born child predicts family succession (a first-born son raises its probability) but has no plausible direct effect on firm performance. Using that instrument, family successions cause operating return on assets to fall by at least 4 percentage points relative to unrelated-CEO successions, with the gap largest in fast-growing industries, industries needing skilled labor, and larger firms. Because the design is quasi-experimental, causal language is warranted here. External validity to large listed firms in other countries is uncertain. Bloom & Van Reenen (2007) add a mechanism from a different angle: among 732 manufacturing firms, family succession by primogeniture — eldest son as CEO — was one of the two main correlates of poor management practices.

RESEARCH FINDING— the counterpoint. Anderson & Reeb (2003), in 403 S&P 500 firms (1992–1999), found founding-family ownership present in about one-third of firms and associated with better accounting and market performance, with family-member CEOs (founder or descendant) outperforming outside CEOs. Endogeneity and survivorship limit causal interpretation, and the later quasi-experimental work finds heir successions specifically hurt. INTERPRETATION The literatures are reconcilable: family ownership — patient capital, long horizons, monitoring — can be an asset, while handing the CEO role to an heir chosen from a pool of one is, on average, costly. Adams, Almeida & Ferreira (2009) add that after correcting for founders' tendency to leave when things go well, founder-CEO leadership in large U.S. firms is associated with, and plausibly causes, better performance. Founders and heirs are not the same case.

RESEARCH FINDING Custódio, Ferreira & Matos (2013) found generalist CEOs — measured by breadth of positions, firms, industries and prior CEO roles — earn a pay premium of about 19% (roughly $1 million per year), largest when firms hire externally, switch from a specialist to a generalist, and assign complex mandates such as restructurings or acquisitions. Pay is a price, not a performance measure. Hambrick & Fukutomi (1991) supply the conceptual clock: CEO tenures pass through seasons — response to mandate, experimentation, selection of an enduring theme, convergence, dysfunction — implying performance tends to rise and then decline over a long tenure, which is why the timing of succession matters and why Shen & Cannella's inverted-U in departing-CEO tenure is not surprising. Miller (1991) found long-tenured CEOs less likely to have organizations aligned with their environments. Wasserman (2003), in 202 Internet start-ups, found that hitting milestones sharply raised the hazard of founder replacement.

Where the evidence is weak

Boards choose outsiders in particular situations, so origin is never randomly assigned; the family-succession instrument is the only clean identification in the set. Samples are modest (109 to 300-odd successions), mostly U.S. large public firms, with accounting-based performance over a few years. "Outsider," "strategic change" and "senior-team turnover" are archival proxies. The seasons model is conceptual. Nothing here tells a board what to do in a specific case; it tells the board which variables to write down.

Why outsiders change more — and why that is not the same as helping

INTERPRETATION Outsiders change more for three reasons that have nothing to do with being better. They have no sunk commitments to the strategy they inherit; a person who did not make a decision finds it easy to reverse. They have no relationships to protect, so the reorganization an insider has avoided for years is, to the outsider, simply Tuesday. And they were hired to change things — the board that goes outside has usually already decided the answer is "different."

Hold that against Zhang & Rajagopalan's (2010) inverted-U. Moderate change helps; excessive change hurts; outsiders amplify both. The board that hires an outsider has bought a higher dose of change and a wider distribution of outcomes. If the company needed a great deal — Karaevli's (2007) poorly performing firm in a turbulent environment — the outsider's dose is medicine. If it needed a little, the same dose is poison, and the poison arrives in the later tenure years, after the honeymoon, when the board has stopped paying close attention.

The three variables

FRAMEWORK Initial disruption, institutional knowledge and mandate decide whether origin helps.

Initial disruption is the cost of a new CEO learning the company: who actually decides things, which numbers are trustworthy, which customers are fragile, which plant manager is holding the place together. An insider has paid this cost already. An outsider pays it in the first year, at the company's expense, and pays more when the senior team is also changing. Shen & Cannella's (2002) turnover interaction is this cost made visible: an outsider who loses the executives who held the institutional knowledge has nothing to integrate into.

Institutional knowledge is the insider's asset and the insider's trap. It makes execution faster and change slower. When the problem is "the strategy is right and the execution is weak," it is exactly what the successor needs. When the problem is "the institution itself is the problem" — the culture, the cost structure, the sacred product line — it is a set of reasons why nothing can change, held by the person in charge of changing it.

Mandate is what the board wants changed, and the variable boards most often leave implicit. "Continue the strategy" and "transform the company" require different people, and a board that cannot say which it wants will hire the candidate who interviews best — which Modules 3 and 14 tell you is not the candidate who executes best.

Followers, contenders and outsiders

Shen & Cannella's (2002) typology is worth carrying around. A follower succeeds a CEO who left on good terms and inherits a mandate of continuity; the risk is too little change. A contender is an insider who succeeds a dismissed CEO — someone who was inside the failure and is now expected to fix it; the risk is that the contender is either implicated in the old strategy or has spent two years positioning against it, and either way the senior team has taken sides. An outsider inherits a mandate of change whether or not the board has said so.

INTERPRETATION The typology explains a common board mistake: dismissing a CEO and then promoting a contender because "we need stability." A dismissal is a mandate for change. A contender who delivers stability has ignored the mandate; a contender who delivers change is, to half the senior team, the person who knifed the old CEO. Shen & Cannella's senior-turnover interaction is where this shows up.

The family case

The family evidence is the most actionable in the module because it is causal. Bennedsen et al.'s (2007) instrument — the gender of the first-born child, which shifts the probability of family succession without plausibly affecting performance — lets us say that choosing an heir over an outside professional lowers operating profitability by around four percentage points on average, most in fast-growing, skill-intensive industries and larger firms. Pérez-González (2006) locates the damage in heirs without an elite education, consistent with a talent pool of one. Bloom & Van Reenen (2007) find primogeniture associated with bad management.

INTERPRETATION None of this says heirs are incapable. It says a selection process with one candidate produces, on average, a worse CEO than a process with many, and the cost rises with the difficulty of the job. Anderson & Reeb (2003) is the reminder that family ownership is a different question. The implication for family boards is not "never appoint an heir." It is "make the heir compete, in a real process, for a mandate you have written down." The family that cannot do that has already decided, and should at least know what the decision costs.

When each makes sense

INTERPRETATION An insider follower makes sense when the strategy is right, the senior team is strong, the departing CEO is leaving in the convergence season rather than the dysfunction season (Hambrick & Fukutomi, 1991), and the problem is execution. An insider contender makes sense when the problem is real but bounded, the contender was visibly outside the failed decisions, and the board will manage the senior-team fallout. An outsider makes sense when performance is poor or the environment turbulent (Karaevli, 2007), when the institution's knowledge is part of the problem, when the mandate is genuinely transformational, and when the company can absorb the integration cost — easier, per Georgakakis & Ruigrok (2017), when the outsider is not demographically alien to the team and brings a broad career. A generalist outsider makes sense for complex mandates where the market already pays for transferable judgment (Custódio et al., 2013). A turnaround specialist makes sense when survival is the problem, with the Module 10 caveat that the same person often misfits the recovery.

The succession-as-fit worksheet

FRAMEWORK Before any candidate is named, the board writes answers to eight questions. The discipline is that questions 1 through 4 must be answered before anyone discusses a name.

1. What is the problem? One sentence. No jargon, no candidate's name. If the board cannot agree on the sentence, it is not ready to choose.
2. What is the mandate? One of: continuity, recalibration, transformation, rescue. Write what "success in three years" looks like in numbers.
3. What dose of change does the company need — and what can it absorb? Where on the inverted-U is the company? What would "too much" look like? Who would leave?
4. What integration cost can the company afford? How long can it run on autopilot while a new CEO learns it? Which three people hold the institutional knowledge, and will they stay?
5. What are the senior-team consequences? For each candidate: who leaves, who is a losing contender, who becomes indispensable.
6. What governance surrounds the successor? Will the board size discretion to the mandate (Module 11)? What is the departing CEO's role — chair, adviser, gone?
7. What is the time horizon, and what are the kill criteria? What evidence, at eighteen months, would mean the board chose wrong?
8. Which candidates fit the answers? Only now: followers, contenders, outsiders (industry, generalist, turnaround), family. Score each against 1–7, not against each other.

The worksheet does not produce an answer. It produces a decision the board can defend and, later, learn from.

Fictional composite.

Halloran Foods was a $650M-revenue packaged-foods company — soups, sauces, a snacks division — with 2,800 employees, listed but 58% owned by the third-generation Halloran family. Frank Halloran, 66, had been CEO for twenty-two years. He was leaving on his own timetable and had told the board he wanted the decision made within six months.

The problem was not dramatic, which made it harder. Private-label competition had taken 400 basis points of gross margin over five years. Two plants needed roughly $90M of automation capex that Frank had deferred. The largest customer was shifting volume to e-commerce, where Halloran's sales organization was weak. Nothing was on fire; everything was slowly cooling.

Three candidates emerged. Maeve Halloran, 41, Frank's daughter, had run the $180M snacks division for four years and grown it 9% annually; she had the loyalty of the plants. Luis Serrano, 58, COO for fourteen years, knew every line and every customer and had deferred the capex alongside Frank. And an outsider: Dana Whitlock, 52, a division president at a $9B food company who had run a margin-recovery program and an e-commerce build-out, and had never worked in a family-controlled business.

The board's first two meetings were about the candidates. The lead independent director, a former CFO named Ruth Adeyemi, stopped the third and asked the board to write the problem in one sentence. It took ninety minutes: Halloran must recover margin through automation and channel shift within four years without losing the plants' loyalty or the family's patience. Mandate: recalibration, not transformation. Dose of change: moderate — the strategy was mostly right; the execution had been deferred. Affordable integration cost: moderate — Serrano and two plant managers held the institutional knowledge and would stay for a follower, probably for Maeve, and might not for Whitlock.

Against that sheet, the picture changed. Serrano was the lowest-disruption choice and the least likely to do what the mandate required, because he had spent fourteen years agreeing not to. Whitlock fit the mandate and carried the highest integration cost — and the board noticed its enthusiasm for her had been partly enthusiasm for a good interview. Maeve fit the family's patience and the plants' loyalty and had not been part of the deferral — but the board had to ask whether it would be considering a 41-year-old with four years of P&L experience if her name were different.

They ran a real process. Maeve was assessed against an external benchmark. Whitlock presented a 100-day plan to the family. Serrano accepted a defined role as president through the capex program regardless of the outcome.

The board chose Maeve, with Whitlock's plan as a template, Serrano as president for three years, and Adeyemi as independent chair with authority to hold the capex timeline. Year two, the automation program ran nine months late because Maeve, under pressure from the plants, slowed it; the board held her to the revised date, and the second plant came in on time. By year four margin had recovered 260 of the 400 basis points. Not the transformation an outsider might have attempted; not the drift a follower might have allowed. A recalibration, delivered by the candidate whose integration cost the company could afford — after the board had checked it was choosing her for the sentence, not the surname.

Put four archetypes in Halloran's chair with the same worksheet.

The Family Business CEO (Maeve) brings identity, patience, the plants' loyalty and the family's tolerance for a four-year program. Gain: the lowest chance of a senior-team rupture and the longest runway. Cost: the risk the family evidence describes — a candidate from a pool of one, whose competence is assumed rather than tested, and whose loyalty to the plants slows exactly the change the mandate requires. The board's job is to make the pool larger than one and the test real.

The Operator CEO (Serrano) is the follower: maximum institutional knowledge, minimum disruption, dials at operational discipline and consensus. Gain: nothing breaks. Cost: nothing changes, because the operator has been part of the decision not to change. This choice makes sense only if the board has decided the deferred capex was right — which it had not.

The Turnaround CEO — an outsider who reads "400 basis points of margin loss" as a crisis — arrives at decisiveness, urgency and centralization. Gain: the automation program happens on time and the sales organization is rebuilt in a year. Cost: Halloran is not in crisis, the dose is wrong for the disease, and Zhang & Rajagopalan's (2010) excessive-change penalty is what this choice buys; the plants' loyalty and two of the three institutional-knowledge holders are the price.

The Visionary CEO — an outsider from adjacent consumer goods who sees Halloran as a brand platform rather than a manufacturer — arrives at innovation, optimism and aggression. Gain: a strategic answer to private label that recalibration cannot provide. Cost: a transformation mandate the board did not write, funded by a family that asked for patience, executed by someone who must first learn which plant manager holds the place together. The mismatch is not the person; it is the mandate.

FRAMEWORK— the Overuse Ladder in succession. Each successor type has a characteristic climb.

The outsider's ladder is adaptability → strategy-of-the-month and decisiveness → impulsiveness. Hired to change things and rewarded early for changing things, the outsider keeps changing things after the company has absorbed what it can. Zhang & Rajagopalan's (2010) finding that the excessive-change penalty is driven by later tenure years is the empirical shape of this ladder: the honeymoon hides it.

The insider follower's ladder is persistence → stubbornness and rigor → bureaucracy: the follower defends the inherited strategy past the point the environment supports it — Miller's (1991) staleness, arriving one succession early. The contender's ladder is dominance → intimidation: having won a contested succession, the contender consolidates power against the losing faction and drives out the institutional knowledge the company needed. The family heir's ladder is confidence → arrogance, with an extra rung: entitlement. The heir who never had to compete does not know what competing would have revealed.

The board's own failure mode is mandate inflation: the mandate written at hiring is "recalibration," but every subsequent board conversation rewards boldness, until the successor is executing a transformation nobody approved. The inverse is mandate deflation: an outsider hired to transform is quietly rewarded for not upsetting anyone.

Early warning signs

• The board discussed candidates before it wrote the problem sentence, or the sentence names a candidate's strength.
• The successor's first-year change dose is larger than the worksheet specified, and the board is pleased.
• Two of the three institutional-knowledge holders have left within eighteen months of an outsider's arrival.
• A contender's first reorganization removes the people who supported the previous CEO.
• A family heir was not assessed against an external benchmark, and the word "obviously" was used.
• Eighteen months in, no one can name the evidence that would have meant the board chose wrong.

1. If you were succeeded tomorrow, write the one-sentence problem your successor would inherit. Now write the sentence your board would write. Are they the same?
2. Which are you: a follower, a contender or an outsider? What mandate did you actually receive, what mandate did you assume, and where did they differ?
3. What dose of change did you deliver in your first two years, and what dose did the company need? Which of the two did you measure?
4. Name the three people who hold your company's institutional knowledge. What would each of them do if an outsider replaced you?
5. If your company is family-controlled: describe the process by which the next CEO will be chosen. How many candidates does it genuinely include?
6. What are the kill criteria for your own tenure — the evidence, at a date, that would mean the board should act? Have you told the board?

• There is no general insider or outsider advantage. Outsiders change more, and strategic change has an inverted-U relationship with performance; outsiders live at both ends of the curve.
• Three variables decide whether origin helps: initial disruption (the integration cost), institutional knowledge (asset or trap, depending on the problem) and mandate (usually unstated).
• A succession is a senior-team event: the consequences of any successor type depend on who else leaves, and losing contenders are either assets or risks depending on what the board does next.
• Family ownership can be an asset; handing the CEO role to an heir chosen from a pool of one lowers profitability on average, by quasi-experimental evidence. Make the heir compete for a written mandate.
• Write the problem, the mandate, the change dose, the integration cost and the kill criteria before discussing a name. The worksheet does not produce the answer; it produces a decision the board can defend and learn from.