Is innovation best managed through contracts or chaos?

Two frameworks for fostering innovation, one from economist Gustavo Manso and the other from author Nassim Taleb, offer competing views on how to structure organizations for breakthrough success.

Where it happened

A June 2026 post on the developer community DEV Community synthesized two influential, and divergent, schools of thought on fostering innovation. The comparison draws on Gustavo Manso's 2011 paper, "Motivating Innovation," and Nassim Nicholas Taleb's 2012 book, Antifragile. While not a direct debate between the two thinkers, the post highlights a fundamental tension in how to manage the uncertainty inherent in creating something new.

Side A: Innovation is a solvable contract problem

Gustavo Manso, an economist at UC Berkeley's Haas School of Business, argues that innovation can be systematically encouraged through carefully designed incentives. The core challenge is the trade-off between exploitation (refining known processes) and exploration (searching for novel solutions). Standard pay-for-performance contracts excel at the former but stifle the latter by punishing the inevitable failures that accompany exploration.

The optimal contract for innovation, according to Manso's model, does the opposite. It combines tolerance for short-term failure with significant rewards for long-term success. This structure provides psychological safety, insulating innovators from pressure to produce predictable, incremental results. It formally protects the right to be wrong on the way to being right. As MIT economist Bengt Holmström, cited by Manso, noted, innovative activities "require an unusual tolerance for failure." Empirical studies appear to support this; one analysis of the venture capital industry found that VCs with higher failure tolerance funded startups that produced more innovative patents.

Side B: Innovation is an emergent property of systems

For Nassim Taleb, focusing on contracts for individual actors misses the point. Innovation is not a manageable process but a feature of systems that are antifragile, meaning they gain from disorder, volatility, and stress. The fragile breaks under pressure, the robust resists it, but the antifragile thrives on it. He argues that top-down, planned research from institutions is ineffective because it tries to eliminate the very randomness that generates breakthroughs.

Taleb’s recipe for funding innovation is what he calls a "1/N" strategy: small amounts per try, many tries, wider than you want. This approach, common among some angel investors and VCs, treats innovation as a portfolio problem. Because the winner in a volatile domain can have an explosive, unlimited payoff, the correct approach is a form of blind diversification. As one venture capitalist told him, "The payoff can be so large that you can't afford not to be in everything." For Taleb, true innovation comes from tinkerers and practitioners taking many small risks, not from theorists operating within a grand design.

What's underneath

The two frameworks are not just different strategies; they operate at different scales. Manso's view is from inside the firm, focused on the principal-agent problem: how does a manager or investor motivate a specific employee or founder? It is a micro-level theory of incentives. Taleb's view is at the system or portfolio level, focused on population dynamics: how does an ecosystem generate positive black swans? It is a macro-level theory of risk and exposure.

Both agree that failure is a necessary input for innovation. The disagreement is over the locus of control. Manso’s work suggests you can architect a local environment for a specific actor to innovate. Taleb’s work suggests that such attempts at control are futile and that the best one can do is structure a portfolio to benefit from systemic randomness, wherever it may strike.

The investor read

This debate mirrors two dominant, viable strategies in venture capital. The 'Manso' model reflects thesis-driven, concentrated funds that provide patient capital. They focus on governance and insulating founders from short-term pressures, believing the right environment can be engineered. The 'Taleb' model reflects the 'spray and pray' portfolio strategy of accelerators and some seed funds. They prioritize broad diversification and deal flow over deep involvement, betting that one outlier will compensate for widespread failure. The continued success of both models suggests they are suited to different stages, founder types, or market conditions, rather than one being definitively superior.