Do Founder-Led Companies Outperform? What the Research Shows

Founder-led companies get a good press. Lists of famous founder-run firms circulate, consultants publish multiples, and investors say they prefer founders at the helm. The honest answer to "do they outperform?" is more qualified. Some well-known studies find a premium. At least one finds none on average. Nearly every study has to deal with a hard problem: founders don't get assigned to companies at random, and the ones who stay in charge are not a typical sample.

This article reviews what the main studies actually say, what each one can and can't tell you, and what seems to change the answer. If you're new to the term, start with what a founder-led company is. For the structural comparison, see founder-led vs professionally managed.

The short version

• There is credible evidence of a founder premium in large US public companies, especially on investment behavior and stock returns in the 1990s.
• The size of that premium depends on the sample, the period, and how "founder-led" is defined. Where a paper gives no number, none is quoted here.
• A study of 94 firms found no average effect on 3-year stock returns, with firm size and age changing the picture.
• Survivorship and selection bias can inflate any "founder-led companies win" claim, and popular multiples rarely explain their method.
• Founders who succeed often get replaced anyway, which is part of why the relationship is hard to measure.

The evidence for a founder premium

Fahlenbrach (2009): investment and stock returns

The most cited paper on the question is Rüdiger Fahlenbrach's "Founder-CEOs, Investment Decisions, and Stock Market Performance," published in the Journal of Financial and Quantitative Analysis (volume 44, issue 2, April 2009, pages 439-466). A university summary of the work, published by Ohio State in 2006, describes a sample of 2,327 large US firms from 1993 to 2002, of which 11 percent were still run by the founder.

The findings, as reported in that summary and the paper's abstract record:

Two things matter when you read these numbers. First, the 4.4 percent figure is the one to remember, because it survives the controls and the 8.3 percent doesn't. Second, this is one period (a single decade in the 1990s) and one population (large US public companies). It says nothing about a 40-person firm, a family-owned manufacturer, or a founder who stays on after a decade of decline.

Adams, Almeida and Ferreira (2009): trying to separate cause from effect

A natural objection to any founder premium is reverse causation: maybe good companies are simply the ones whose founders stay. Adams, Almeida and Ferreira took this on in "Understanding the relationship between founder-CEOs and firm performance," in the Journal of Empirical Finance (volume 16, issue 1, pages 136-150). Their abstract describes the method: instrumental variables, using the proportion of a firm's founders who are dead and the number of people who founded the company.

Their results, in direction only:

• Founder-CEO status is endogenous in performance regressions, meaning it can't be treated as a clean input.
• Good performance makes it less likely that the founder keeps the CEO title.
• After factoring that out, they report a positive causal effect of founder-CEOs on performance, larger than the standard regression estimate.
• Founder-CEOs are more likely to give up the CEO post after unusually low or unusually high performance.

That last point cuts both ways and is worth sitting with. It means a simple snapshot of "which companies currently have founder-CEOs" already filters out some of the best and some of the worst performers.

Villalonga and Amit (2006): founders versus descendants

Belén Villalonga and Raphael Amit's "How do family ownership, control and management affect firm value?" (Journal of Financial Economics, volume 80, issue 2, pages 385-417, 2006) is about family firms, not founder firms as such, but it contains the clearest evidence on a related question: does it matter who leads?

Wharton's summary of the paper says the authors analyzed proxy data on Fortune 500 firms from 1994 to 2000. Family firms did better than non-family firms when the founder served as CEO, or as chairman with a hired CEO. When a descendant served as CEO, even with the founder as chairman, firm market value declined. The takeaway for this article is that the premium, where it exists, attaches to the founder and not to family control in general. For the succession side of that story, see family business succession and the non-family CEO.

Nelson (2003): the market likes founders at the IPO

Nelson's "The persistence of founder influence: management, ownership, and performance effects at initial public offering" (Strategic Management Journal, 2003) looked at firms at the point of going public. Its abstract reports that founder influence persists in governance and ownership arrangements, and that the stock market reaction to founder-led firms at IPO was higher than for the comparison group, relative to accounting value. That's a statement about investor expectations at one moment, not about long-run results.

The popular number: Bain's 3.1x

The most repeated figure comes from Bain & Company, in a piece published on Bain's site on 23 March 2016 and originally in Harvard Business Review. It says an index of Fortune 500 companies in which the founder is still deeply involved performed 3.1 times better than the rest over the past 15 years. The same article reports a study by three Purdue professors finding that S&P 500 companies where the founder is still CEO generate 31 percent more patents.

Treat the 3.1x with care, for three reasons visible on the page itself:

1. The definition is broad. The article describes the relevant founders as playing a significant role as CEO, chairman, board member, owner, or adviser. That's a very different group from "founder-CEO."
2. The method isn't described. The page doesn't say how the index was built, how companies were selected, or how the 15 years were measured.
3. It names winners. The examples given (Oracle, Haier, L Brands, Four Seasons) are companies that stayed successful with a founder around, and the article doesn't discuss how failed founder-led firms are counted.

It's a useful hypothesis from a consultancy that sells advice built on it. It isn't peer-reviewed evidence, and it shouldn't be quoted as a general law.

The evidence against a simple premium

Jayaraman and colleagues (2000): no average effect

In "CEO founder status and firm financial performance" (Strategic Management Journal, volume 21, issue 12, pages 1215-1224, 2000), Jayaraman, Khorana, Nelling and Covin note that earlier research had produced inconsistent results. Their own study of 94 founder- and nonfounder-managed firms found that founder management had no main effect on stock returns over a 3-year holding period, but that firm size and firm age moderate the relationship.

The sample is small and the paper is old, so it shouldn't be treated as the last word. Its value is as a reminder that "founder premium" isn't a settled constant. The abstract says the effect depends on context, and it names two contextual variables.

Wasserman (2003): why founders get replaced

Noam Wasserman's "Founder-CEO succession and the paradox of entrepreneurial success" (Organization Science, volume 14, issue 2, pages 149-172, 2003) studied the first replacement of a founder-CEO in 202 Internet start-ups. It identifies two events that affect the rate of founder replacement: completing product development and raising each round of outside financing. The paper's term for the pattern is "paradoxes of success": reaching the milestones a founder was supposed to reach changes the odds that the founder keeps the job.

This isn't a performance study. It helps explain why founder-led firms at later stages are a filtered group. Outside investors often have reasons to bring in other leaders as a company matures, and where they do, the remaining founder-CEOs aren't a random draw. For the other side of the coin, where a founder holds on for too long, see founder's syndrome.

Survivorship bias and selection effects

These are the two issues that most often get lost in popular summaries.

Survivorship bias happens when you study the companies that exist now and ignore those that didn't make it. If you list large companies that are still founder-led today, you are looking at firms that survived long enough to be large. Founder-led companies that failed, were sold, or replaced the founder aren't on the list. A claim like "founder-led companies beat the market" can be true of the visible survivors and false of all founder-led companies that were ever started.

Selection effects work in several directions at once:

• Good firms keep founders. Adams, Almeida and Ferreira found endogeneity in founder status, so a correlation between founder-CEO status and performance doesn't prove the founder caused the result.
• Good firms lose founders. The same paper found founders are more likely to leave the CEO role after unusually high performance as well as after unusually low performance. A sample that excludes departed founders loses some of the strongest cases.
• Large public firms are a special group. Fahlenbrach's and Bain's samples are big, listed companies. A founder who can run a Fortune 500 firm may be exceptional. Results for them don't carry over to a 50-person private business.
• Definitions vary. "Founder-led" can mean founder-CEO, founder on the board, or founder as a significant owner. A study using one definition can't be compared with a study using another.

None of this means the premium is imaginary. Adams and colleagues' instrumental variables approach is an attempt to deal with exactly these problems, and it still reports a positive effect. But it does mean that anyone who quotes a single multiple with no sample, definition, or period is giving you less than they appear to.

What moderates the effect

Read together, the studies point to a handful of factors that change the answer. These are patterns suggested by the papers above, not settled laws.

What this means for a leadership team

If you run, invest in, or work for a founder-led business, the research supports a few cautious conclusions:

1. Don't assume the premium is yours. The strongest evidence is for large public companies in a particular period. Your firm's size, age, and sector all matter.
2. Treat founder involvement as a different thing from founder control. A founder-CEO, a founder-chair with a hired CEO, and a founder who is simply a large shareholder are three different arrangements, and the studies don't treat them alike.
3. Look at what the premium seems to work through. Where there's a gain, the cited mechanism is investment behavior: more R&D, more capital expenditure, more focused acquisitions. That's something a company can ask about directly instead of relying on a label.
4. Plan for the transition regardless. Whether or not a premium exists today, founders eventually step back. Succession planning, professionalizing the business, and managing key person risk are how a company keeps what's good about founder leadership while reducing dependence on one person.

Key Facts: Founder-led company performance research

• Fahlenbrach studied 2,327 large US firms from 1993 to 2002; 11 percent were led by the founder.
• The same study reports a benchmark-adjusted return of 8.3 percent a year for an equal-weighted founder-CEO strategy, and 4.4 percent a year after controls (abstract record).
• Adams, Almeida and Ferreira report a positive causal effect of founder-CEOs after addressing reverse causation, and find founders more likely to leave the CEO post after unusually low or high performance.
• Villalonga and Amit (Fortune 500, 1994 to 2000): value is higher with a founder-CEO or founder-chair, and falls with a descendant CEO.
• Jayaraman and colleagues found no main effect of founder management on 3-year stock returns in 94 firms, with size and age as moderators.
• Bain's 2016 article cites a 3.1x index result for Fortune 500 firms with deeply involved founders, without describing its method on the page.

Related reading

• What Is a Founder-Led Company?
• Founder-Led vs Professionally Managed
• Founder's Syndrome
• Key Person Risk
• Professionalizing a Business
• Succession Planning