Monkeys, Trees, and Economic Complexity

Harvard economist Ricardo Hausmann has a vivid way of explaining economic development: imagine products as trees in a vast forest, companies as monkeys, and countries as collections of monkeys distributed across different trees. Economic development means monkeys jumping from poorer parts of the forest to richer parts.

The key insight? Trees that are close together (products requiring similar capabilities) allow easy movement. Trees far apart make jumps nearly impossible. Dense parts of the forest—machinery, chemicals, electronics—enable rapid capability expansion. Sparse parts—raw materials, basic agriculture—trap countries in low-complexity activities.

This "product space" model reveals why the global textile mill problem from Part 1 persists, and more importantly, how to solve it through complementary rather than competitive development.

The Knowledge Transfer Revolution

The most successful economic development stories aren't about countries competing for the same trees—they're about strategic knowledge transfer that helps nations build complementary capabilities. The semiconductor industry provides the perfect example.

When Taiwan Semiconductor Manufacturing Company (TSMC) was founded in 1987, it received crucial technology transfer from Philips, which also invested 27.5% of the initial capital. This wasn't charity—it was a strategic complement. Philips wanted to focus on chip design and applications (high-value activities) while TSMC specialized in advanced manufacturing (a different but equally valuable capability).

The result? Both parties won. TSMC became the world's leading foundry, while Philips concentrated on higher-value design work. This created a complementary relationship that enabled the entire semiconductor ecosystem to flourish rather than just redistributing existing value.

The lesson: successful knowledge transfer creates new capability trees rather than just moving monkeys to existing ones.

Why Most Technology Transfer Fails

Traditional development models often fail because they try to replicate rather than complement existing capabilities. When every country attempts to build Silicon Valley, they're essentially putting all the monkeys on the same tree branches, driving down returns for everyone.

The failure pattern is predictable:

1. Developing country targets high-value industry (semiconductors, biotech, aerospace)

2. Attempts to replicate entire value chain domestically

3. Lacks complementary capabilities (IP frameworks, specialized suppliers, risk capital)

4. Results in low-quality imitation rather than innovation

5. Competes on cost rather than value, recreating the textile mill problem

Better approach: identify adjacent capabilities where the country can become world-class while remaining complementary to, rather than competitive with, existing leaders.

The Aerospace Model: Collaboration Over Competition

Quebec's aerospace cluster demonstrates how industries can shift from zero-sum competition to positive-sum collaboration. Instead of trying to build vertically integrated aerospace giants to compete with Boeing, Quebec developed specialized supplier capabilities that made the entire North American aerospace industry more competitive.

Key elements of success:

• Specialized excellence: Focus on specific capabilities (avionics, materials, precision manufacturing) rather than complete systems

• Collaborative governance: Industry associations that balance competition with cooperation

• Shared infrastructure: Joint research facilities, testing centers, training programs

• Long-term relationships: Partnerships that create mutual dependencies rather than arm's-length transactions

The result is an aerospace ecosystem where Canadian suppliers, American system integrators, and global customers all benefit from specialized complementarity.

Building Complementary Capability Trees

The monkeys-and-trees model suggests three strategic approaches for escaping the textile mill trap:

1. Vertical Specialization

Different countries focus on different stages of the value chain:

• United States: Fundamental research, platform technologies, brand development

• Germany: Advanced manufacturing systems, precision engineering, industrial automation

• Taiwan: Semiconductor fabrication, advanced materials, process optimization

• India: Software services, data analytics, technical support

Each country becomes world-class in its specialization while remaining complementary to others.

2. Horizontal Complementarity

Countries develop capabilities in related but non-competing product spaces:

• Switzerland: Precision instruments, luxury goods, pharmaceutical research

• Denmark: Renewable energy systems, agricultural technology, logistics innovation

• Singapore: Financial services, supply chain management, urban technology

These capabilities reinforce each other without direct competition.

3. Temporal Laddering

Advanced economies continuously move to newer technologies while transferring mature capabilities:

• Semiconductors: U.S. develops 3nm chips while transferring 14nm technology

• Automotive: Germany advances electric vehicle systems while transferring combustion engine expertise

• Pharmaceuticals: Switzerland focuses on breakthrough drugs while transferring established manufacturing processes

This creates a ladder of development where everyone can climb without pushing others down.

The Financial Architecture Problem

Current global financial flows work against complementary development. Manufacturing nations accumulate surpluses that get recycled into U.S. Treasury bonds, essentially financing American consumption rather than building domestic capabilities.

As investment banker Scott Bok revealed, this system emerged because U.S. financial institutions didn't just move capital—they systematically embedded American business practices globally while retaining control over high-value financial services. The result is a financial architecture that channels global savings toward American assets rather than distributed capability building.

Reform requires redirecting these flows:

• Innovation investment channels: Direct surplus savings toward R&D partnerships rather than Treasury bonds

• Capability building funds: Support domestic brand development and service sector growth in surplus countries

• Collaborative industrial policy: Sectoral agreements that enhance rather than compete with each nation's strengths

From Remote Work to Distributed Innovation

The COVID-19 pandemic accidentally demonstrated how to escape traditional brain-drain patterns. Programs like Tulsa Remote successfully recruited 3,300 remote workers, showing that innovation doesn't require expensive coastal concentration.

This opens possibilities for distributed innovation networks:

• Regional innovation centers: Shared facilities connected through knowledge networks rather than geographic proximity

• Virtual collaboration platforms: Model-based systems engineering enables distributed design while maintaining IP security

• Talent circulation: Professionals can contribute to global innovation while living in lower-cost regions

The key insight: innovation happens in networks, not just places. When knowledge work becomes truly distributed, countries can build innovation capabilities without recreating Silicon Valley's cost structure.

Practical Implementation

Moving from theory to practice requires specific mechanisms:

Graduated Technology Transfer

• Start with mature technologies where IP protection is less critical

• Build institutional frameworks (IP protection, quality standards, dispute resolution)

• Gradually transfer more advanced capabilities as trust and competence develop

• Maintain control over core platform technologies while enabling local adaptation

Complementary Research Programs

• Joint research facilities focused on different aspects of the same challenges

• Cross-licensing agreements that create mutual dependencies

• Shared standards development that benefits entire ecosystems

• Risk sharing for breakthrough technologies that no single country can finance

Balanced Development Compacts

• Formal agreements linking export growth to domestic consumption growth

• Requirements for local brand investment alongside production capacity

• Minimum wage growth tied to productivity gains in export sectors

• Service sector development incentives that reduce export dependency

The Network Effect

The most powerful aspect of complementary development is that it creates network effects rather than zero-sum competition. When Taiwan excels at semiconductor fabrication, it makes American chip designers more valuable, not less. When Germany develops advanced manufacturing systems, it enhances rather than threatens American innovation capabilities.

This is the opposite of the textile mill dynamic, where improvements by one player reduce returns for everyone else.

Complementary capabilities create virtuous cycles:

• Specialization enables world-class excellence

• Excellence creates premium pricing power

• Premium pricing funds further innovation

• Innovation enhances the entire network's competitiveness

Coming Next

In Part 3, we'll examine the accounting reforms needed to make this transition politically feasible. Current trade statistics obscure complementary relationships while exaggerating competitive tensions. Better metrics would reveal that successful economies are already more complementary than competitive—we just can't see it through our 19th-century accounting frameworks.

We'll show specifically how reformed trade accounting would reduce measured deficits while incentivizing exactly the kind of complementary development outlined here. The goal isn't just better bookkeeping—it's creating transparency that enables better policy decisions.

Because you can only optimize what you can measure, and we're currently measuring the wrong things.

Sources and Further Reading

Economic Complexity Theory:

• Hausmann, R., Hidalgo, C., et al. (2011). The Atlas of Economic Complexity. Harvard CID

• Hidalgo, C. & Hausmann, R. (2009). "The Product Space Conditions the Development of Nations." Science

• Economic Complexity Observatory: atlas.cid.harvard.edu

Knowledge Transfer Case Studies:

• Mathews, J. & Cho, D. (2000). "Tiger Technology: The Creation of a Semiconductor Industry in East Asia." Cambridge University Press

• Spencer, J. (2008). "The Global Expansion of Taiwan's Semiconductor Industry." Research Policy

• Macher, J., Mowery, D. & Hodges, D. (1999). "Semiconductors and the Internet." U.S. Industry in 2000

Aerospace Industry Evolution:

• Niosi, J. & Zhegu, M. (2005). "Aerospace Clusters: Local or Global Knowledge Spillovers?" Industry and Innovation

• Quebec Aerospace Association: aeroqc.ca

• International Association of Science Parks: iasp.ws

Distributed Innovation Research:

• Florida, R. (2017). The New Urban Crisis. Basic Books

• Moretti, E. (2012). The New Geography of Jobs. Houghton Mifflin Harcourt

• Tulsa Remote Program Results: tulsaremote.com/about

Technology Transfer Frameworks:

• Archibugi, D. & Michie, J. (1995). "The Globalisation of Technology: A New Taxonomy." Cambridge Journal of Economics

• UNCTAD World Investment Report 2005: "Transnational Corporations and the Internationalization of R&D"

Model-Based Systems Engineering:

• INCOSE Systems Engineering Handbook v4.0

MIT Systems Engineering Program: web.mit.edu/syeng