Growth markets are node-scarce, not idea-scarce

Authored by Shamil Ibragimov, Scholar-in-Residence

MIT Kuo Sharper Center for Prosperity and Entrepreneurship 

For years, physicists tried to answer a deceptively simple question: why do some things like cities, ideas, innovation ecosystems grow disproportionately larger and more influential than others? Why do a few nodes attract attention, resources, and talent, while most remain peripheral?

The answer, it turns out, has little to do with fairness and a lot to do with how networks work. In the late 1990s, physicist Albert-László Barabási discovered that many real-world systems, from the internet to scientific collaboration, share a common structure. They are what he called “scale-free” networks. Networks in which most nodes have very few connections, while a small number become highly connected hubs, with no typical or “average” level of influence. These networks are governed by two universal laws: growth and preferential attachment. Networks start small, expand by adding new nodes, and those new nodes tend to connect to actors who are already well connected. Success, visibility, and connectivity compound.

This insight emerged not from economics or public policy, but from physics. Yet it helps explain one of the most persistent puzzles of innovation in growth markets: why these societies are often overflowing with ideas, talent, and ambition, and yet struggle to translate them into sustained innovation. The problem is not a lack of ideas. It is a lack of nodes - people, institutions, and places - that are sufficiently connected to turn ideas into action.

Innovation ecosystems are usually described in institutional terms of ministries, laws, funds, incubators, technology parks and stakeholders. But Barabasi’s work, popularized in his book Linked, shows that networks organize themselves around people and relationships long before they solidify into institutions. Institutions do not create networks, networks create institutions. But it’s also important that those nodes activate specific mechanisms for connecting the nodes to each other, for proactively orchestrating levers that will push early stage and growth stage initiatives to the next level.

In growth markets, public policy for innovation frequently gets this sequence backwards. Governments focus on building organizations without ensuring that those organizations become attractive hubs in a living network. Funding programs are launched without credible actors to anchor them. Innovation strategies assume linear causality: build the institution, and innovation will follow. But networks do not grow linearly. They grow unevenly, clustering around visible, trusted, well-connected nodes.

From a network perspective, this does not imply that innovation policy is designed to privilege a few actors over others. Rather, it helps explain an observable and persistent pattern: why influence, talent, and capital tend to accumulate around already well-connected people and places. Preferential attachment is not a moral choice or a policy bias; it is a structural property of growing networks. New participants gravitate toward visible hubs because doing so reduces uncertainty, frictions and increases their chances of success.

This is one reason Silicon Valley, despite repeated predictions of its decline, remains the most attractive destination for startups globally. Its advantage lies not only in access to capital or technology, but in the density of connections among entrepreneurs, investors, researchers, and experienced operators. For newcomers, attaching to this network is rational, because of the proactive nature of the players that recommend resources to each other. Past success becomes a signal and that signal reinforces itself.

Why nodes behave the way they do

Network science alone, however, risks sounding deterministic, as if individuals were merely particles pulled by invisible forces. This is where economist Ronald Coase sharpens the picture.

In his seminal essay The Nature of the Firm, Coase asked a simple question: why do firms exist at all? His answer was grounded in human behavior. Firms emerge when they reduce transaction costs, uncertainty, negotiation, coordination, compared to relying on the market.

Applied to innovation ecosystems, the lesson is straightforward. Participation is rational. Individuals choose where to invest their time, talent, and capital based on incentives, trust, status, and friction. They do not join networks because a policy document tells them to. They join when the cost of participation is low and the potential upside like economic, reputational, or intellectual is high. This matters because policies that ignore incentives cannot shape networks, no matter how well intentioned they are. 

Nodes, in other words, are not abstract system elements. They are people with motives, cultures, assumptions, and constraints. Institutions are simply the temporary crystallization of those human choices. This is why leadership matters so much in innovation ecosystems. Individuals who can bridge social worlds like scientists and entrepreneurs, local founders and global capital, government and private initiatives, often become outsized nodes themselves. They bind communities together and make collective action possible.

What growth markets often get wrong

Many innovation policies in growth markets rest on three flawed assumptions.
The first is linear causality: the belief that inputs reliably produce outputs. Build a technology park, fund a startup program, and innovation will follow. But ecosystems are not assembly lines.

As JF Gauthier, CEO of Startup Genome, once recounted in an interview with Tech Ukraine, Singapore’s early attempts to replicate Israel’s success illustrate this point well. In the mid-1990s, Enterprise Singapore adapted the Yozma model almost verbatim. On paper, the institutions were sound. In practice, the impact was limited. What was missing was not policy design, but cultural readiness: a society still oriented toward stable employment, low risk tolerance, and predictable career paths. Before venture capital could work, an entrepreneurial community had to emerge. The lesson is not that models cannot travel, but that ecosystems do not respond to policy inputs in isolation. Without attention to culture, incentives, and social norms, even well-designed interventions struggle to take root.

The second is a focus on institutions without networks. Buildings rise, regulations are passed, funds are allocated and yet the human connections that give these structures life remain thin.

The third is funding programs without creating attractive nodes. Resources are spread widely in the name of fairness, rather than concentrated where they can generate visibility, credibility, and momentum.

A good public policy for innovation does something different. It reduces friction within nodes, making it easier for entrepreneurs, researchers, and investors to work together, and reduces friction between nodes, so ideas, talent, and capital can move quickly across boundaries. Its aim is not equality of distribution, but connectivity and attraction.

When we talk about innovation, we often default to startups or companies. But innovation is not fundamentally about firms. It is about the right relationships among the right people at the right time, allowing new ideas to take whatever organizational form they need. Companies are one possible outcome, not the starting point. From this perspective, the superficial manifestations of innovation policy as the number of startups, incubators, or grants are secondary. The primary unit of analysis is the quality of human relationships. It’s the willingness of those humans in those relationships to advocate on behalf of each other, join forces to inspire institutions to make the changes required for businesses/industries to scale, and the proactive participation of those humans.

There is a reason we borrow the biological term “ecosystem” to describe innovation. Natural ecosystems are diverse, complex, and unpredictable. They depend on the interaction of many elements like plants, animals, fungi, microbes and none of which alone determines the outcome. That complexity creates serendipity: unexpected discoveries, adaptations, and breakthroughs.

Innovation ecosystems work the same way. They are not farms, where inputs can be measured against outputs with precision. It is not enough to have the right ingredients; what matters is how they interact, and under what conditions.

Several years ago, friends of mine, successful entrepreneurs from Almaty, Kazakhstan, traveled to southern Kyrgyzstan. There, over lunch, they fell in love with a local specialty: tandyr samsa, a rich meat-filled pastry baked in a clay oven. Inspired, they decided to recreate it in Almaty. They built an identical oven, used nearly the same ingredients, and even hired the same chef. Yet the result tasted completely different.

Only later did they realize what was missing. In southern Kyrgyzstan, the ovens are fired with cotton stalks, an agricultural byproduct that costs nothing locally and gives the food its distinctive aroma. Transporting those stalks hundreds of miles made no economic sense. The flavor belonged to the place.

Innovation ecosystems work the same way. You can copy visible structures, import expertise, and replicate institutions and still miss the essence. Context matters, not as an excuse for underperformance, but as a source of uniqueness. The task of public policy is not to predetermine which sectors will succeed, but to create dense, credible nodes and bring the right players together. Once those connections exist, participants tend to discover for themselves where their comparative advantages lie, whether in Fintech, health tech, climate tech, or domains that policy designers could not have anticipated.

When nodes form organically, they do more than support individual enterprises. They accelerate investment across the ecosystem, lower coordination costs, and enable collective action. Over time, these networks begin to articulate their own needs, aligning entrepreneurs, investors, and researchers around shared priorities. It is through this process that institutions are shaped from the bottom up, and public policy evolves in response to real economic activity, helping regions build innovation engines with distinct, locally grounded strengths rather than imitating models developed elsewhere.

There is no universal recipe for building innovation ecosystems. Every place requires its own approach to accelerating node formation and strengthening networks. What is universal is the need for clarity of purpose and a human-centric approach to public policy.

Growth markets do not lack ideas. They lack dense, credible, and connected nodes that allow ideas to travel, collide, and compound. The task of public policy is not to manufacture innovation directly, but to shape the conditions under which networks can grow, and people can choose, rationally and enthusiastically, to join them.