Why Unequal Equity Splits Might Actually Help Your Startup Innovate Better

Mar 24, 2026

This is a practical review of the academic article “Promoting novelty creation: The role of ownership distribution in new venture teams” by Elisabeth Mueller, Laura Bregenzer, Patrick Figge, and Carolin Haeussler, (soon to be) published in Strategic Entrepreneurship Journal (SEJ)—one of the top peer-reviewed academic entrepreneurship journals.

I wrote this review—to be followed by others—to help bridge the gap between academic research and practice in entrepreneurship.

------------------------------------------ 

There are two schools of thought on how startup founders should split equity. One, represented in practice by Y-Combinator in the US, suggests an equal split facilitates long-term collaboration grounded in fairness. The other, represented by the Venture Capital (VC) investor Earlybird in Europe, supports the idea that clear leadership bolstered by unequal ownership improves a startup’s ability to innovate and scale(1).

Mueller and colleagues address this issue using a dataset most US-based researchers would envy: the Mannheim Enterprise Panel, covering nearly the entire population of German firms. Germany requires limited liability companies (Gesellschaft mit beschränkter Haftung, GmbH) to disclose owner names and shareholding in a public registry. So, instead of relying on voluntary surveys, the authors use ownership data for over 2,000 team-founded firms and over 5,000 patents between 1999–2015. They match this to patent records from PATSTAT from the European Patent Office (EPO) and supplement it with 40 interviews. Their main dependent variable (i.e., the outcome in which they try to explain the observed variance) is Recombinant Novelty(2)—which they operationalize with the count of new pairwise combinations of technology classifications appearing on a patent document for the first time. Think of it as a proxy for ex-ante novelty: the more new pairs, the higher the novelty.

Main Findings

Unequal equity splits are associated with higher novelty. Patents from firms with unequal ownership among co-founders had meaningfully higher recombinant novelty than those with equal splits. In other words, uneven ownership splits among founding team members is associated with more innovative patents.

Majority ownership matters most. Among unequal teams, those with a single majority owner (>50%) had the highest novelty in their patents. Teams where everyone was a minority owner showed a positive but statistically weak effect. So, the effect appears not to be simply due to the inequality in shareholding; but it seems to be more due to having a clear majority owner.

When founders from unequal shareholding teams get involved in R&D, novelty goes up—regardless of their personal share. I think this is the most interesting finding. As individuals, both majority and minority owner-inventors from teams with unequal equity holdings produced more novel patents compared to shareholders with equal ownership. Surprisingly, though, the team’s ownership structure seems to matter even more than the individual inventor’s equity stake. This is indeed interesting.

Interviews revealed different decision-making cultures. Equal equity teams described slow, consensus-based processes leading to “the lowest common denominator.” While this might be a good thing in most contexts, it is not ideal when the focus is on generating highly novel innovations that can give an edge to a startup. Members of unequal shareholding teams described a process of guided decision-making where even minority owners had substantial autonomy within their domains. [One interviewee from an equal shareholding team said: “We are all exactly the same and we decide everything together. And that is very naïve retrospectively, because that involves a lot of time and coordination.” (p. 20.) A remarkably candid and insightful comment, IMHO.]

The Academic Researcher View of the Paper

The paper’s greatest strength is its data—verified, longitudinal, and objective ownership data at the individual level, covering nearly the entire population of German firms. This is golden! The abductive approach used in research design—evaluating four competing explanations instead of testing pre-specified hypotheses—is also refreshing to see in an academic paper that addresses a practical issue like this. So, one could say that the novelty of the paper comes from combining these two existing elements to achieve high recombinant novelty itself! [Sorry, I could not resist.]

My major methodological concern, which the authors also acknowledge, is endogeneity. The formation of entrepreneurial founding teams is characteristically an endogenous affair, unlike most other work teams(3). Founders who might have gravitated toward unequal splits might have done so because of an underlying variance in their ambition towards and/or desire for control of the venture, with the same variance possibly driving the push for bolder innovation. The authors use omitted-variable diagnostics and firm fixed-effects, suggesting results are robust to moderately strong confounding factors, but this is not the same as ‘evidence for a causal relationship’. As such, the paper avoids claiming causality, but I don’t see this as a problem. A lot of stuff in life is correlational (= hard to ascertain causality with the evidence and methods of inquiry available to us), and knowing more about what correlates with what, and how we can use that information to make decisions of consequence is valuable. The paper surely makes a contribution to our knowledge in this regard.

A second concern is generalizability. Ownership-based entrepreneurship with long-term shareholding intentions is deeply rooted in Germany’s Mittelstand culture. This is quite different from Silicon Valley’s entrepreneurial ecosystem where fluid founding teams and use of preferred shares or complex option vesting schedules are common. What “majority owner” means in a German GmbH vs. a Delaware C-Corp are very different things(4), with possibly divergent long-term implications for startups.

Takeaways for Founding Teams & Investors

Equity split is a governance decision, not just a financial one. How you divide equity shapes how your team makes decisions under uncertainty. Equal splits may feel fair, but they can also create consensus traps where bold ideas get whittled down to the most acceptable ones. If innovation is essential for your venture, consider whether your ownership structure—even if implemented with the best of intentions—might be inadvertently holding back your startup’s full innovative potential.

A clearly designated decision-maker can empower everyone. A counterintuitive finding of the study is that even minority owners in unequal shareholding teams were more innovative, individually, than owners in equal teams. A designated owner-leader was conducive for everyone to act more autonomously, avoiding analysis paralysis. This is consistent with what organizational researchers have long known about clear authority structures reducing coordination costs (e.g., Aghion & Tirole, 1997)(5). Give someone the mandate to push through uncertainty, and, in all likelihood, that would benefit all founding team members.

If you’re a technical founder, continue to be involved in R&D. Founders from unequal-ownership teams who participated directly in inventive work produced more novel patents. Stepping entirely into a management role (or being pulled/pushed into one) prematurely may hurt the firm’s innovative output.

In Sum

The paper leaves some questions open, such as potential downsides of concentrated ownership for team cohesion, and implications for minority founder retention/exit following intra-team conflict. But the richness of the data makes one want to ask those kinds of questions, which is a compliment to a study that is well conducted, with plenty of contributions to both the academic research on and the practice of entrepreneurship. I look forward to more from the authors in follow-up studies.

------------------------------------------ 

Disclaimer: I used Claude (Opus 4.6) to edit this post. All ideas are mine.

------------------------------------------ 

(1) This is not an official position, but inferred from an Earlybird co-founder’s 2024 podcast remarks suggesting that clarity in leadership has been essential for their portfolio firms’ innovation and scaling (adapted from the Mueller et al., 2026 paper reviewed here).

(2) The Schumpeterian view is that innovation results from combining existing ideas in new ways. These knowledge combinations can lead to major novelties in technology and the economy, or they can manifest as relatively minor improvements in various uses, features, or components of products/services/systems in existence—a concept referred to as radical vs. incremental innovation. Recombinant novelty refers to the “level of newness” of an innovation, which helps convert the radical-incremental dichotomy into a continuum of innovation novelty. Example: the first iPhone was highly novel compared to button/flip phones like the Nokia 3310; but recent iPhone models are mostly incremental compared to their immediate predecessors. So, the first iPhone can be said to have a higher degree of recombinant novelty, in comparison to the more recent iPhone models.

(3) Lazar, M., Miron-Spektor, E., Agarwal, R., Erez, M., Goldfarb, B., & Chen, G. (2020). Entrepreneurial Team Formation. Academy of Management Annals, 14(1), 29–59. https://doi.org/10.5465/annals.2017.0131

(4) Maybe both Y-Combinator and Earlybird are doing what’s most practical in their respective ecosystems. In Europe a dominant shareholder amongst founding team members might improve innovation outcomes through clear leadership signals and well-articulated roles in innovation workflows, while in the US the messiness of this relationship might make a deliberate choice moot, leading to the ‘equal split is best' heuristic.

(5) e.g., Aghion, P., & Tirole, J. (1997). Formal and real authority in organizations. Journal of Political Economy, 105(1), 1–29. https://www.journals.uchicago.edu/doi/abs/10.1086/262063