Supply-Chain Diversification vs Sector and Country Diversification
July 5, 2026
Altsets
Research by Altsets Research
Compare sector and country diversification with supplier, customer, end-market, and network overlap so globally distributed holdings are not assumed to be economically independent.
Data used:Altsets Supply Chain Intelligence: 90k+ entities, 400k+ relationships, 20+ years of history.
Key findings
- The supplied network contains a connected ASML to Micron to Nvidia to Quanta path spanning the Netherlands, the United States, and Taiwan across several business functions.
- Sector and country diversification can therefore coexist with common supplier, customer, end-market, or network exposure, while direct relationship percentages should not be multiplied across hops.
A portfolio can be diversified by sector and country while remaining concentrated in one supply-chain demand cycle. The supplied Altsets network provides a concrete example. A visible path connects ASML to Micron to Nvidia to Quanta Computer.
The companies span the Netherlands, the United States, and Taiwan. They also occupy different economic roles: semiconductor equipment, memory, accelerated computing, and server or data-center systems. Those differences can make a portfolio look diversified using conventional labels. The relationship network shows that the companies can still participate in one connected AI-infrastructure chain.
Sector diversification answers a classification question
Sector diversification asks whether a portfolio is spread across different industries. That is useful. It reduces exposure to some industry-specific risks.
But sector classifications do not describe who buys from whom. Two companies in different sub-industries can still depend on the same customer or capital-spending cycle.
Country diversification answers a jurisdiction question
Country diversification can reduce exposure to one legal, political, currency, or macroeconomic environment. It does not guarantee economic independence. The ASML-Micron-Nvidia-Quanta path spans several countries.
A global AI infrastructure slowdown could still affect several nodes in the path even if no single country shock occurred. Country and supply-chain diversification therefore measure different things.
The actual relationships show the common chain
In the displayed data:
- ASML to Micron is associated with 7.64% of ASML revenue and 11.91% of Micron's cost base.
- Micron to Nvidia is associated with 17.62% of Micron revenue and 14.00% of Nvidia's cost base.
- Nvidia to Quanta is associated with 0.53% of Nvidia revenue and 1.76% of Quanta's cost base.
Those percentages should not be multiplied across the path. They show that each direct relationship has its own economic significance. The path itself reveals connected exposure.
A four-stock portfolio can still share one economic driver
Imagine a portfolio holding ASML, Micron, Nvidia, and Quanta. The holdings differ by geography and business function. Yet the network suggests a common dependency on continued semiconductor and data-center investment.
That does not mean all four stocks move together every day. ASML depends on equipment demand across many customers. Micron has customers beyond Nvidia.
Nvidia serves multiple markets. Quanta builds systems for more than one customer. Supply-chain diversification is not about declaring the holdings identical. It is about measuring one source of common economic exposure that sector and country labels do not capture.
The same network can reveal genuine diversification too
Supply-chain analysis is not only a concentration detector. Two holdings can sit in the same sector but depend on different suppliers, customers, geographies, or end markets. That can create more economic diversification than a sector label suggests.
The goal is not to replace sector diversification. It is to add another dimension.
Supply-chain diversification needs several layers
A useful portfolio review can ask whether holdings share the same large customers, the same critical suppliers, the same second-order bottlenecks, the same manufacturing geography, the same end-market demand, or the same infrastructure buildout cycle. Each overlap answers a different question. A simple count of shared company names is only the beginning.
Structural relationships can still reveal overlap
Not every useful edge needs a complete set of metrics. A structural relationship can show that two holdings connect to the same counterparty or network cluster. That is enough to flag a diversification question.
Quantified metrics determine whether the overlap appears economically large enough to prioritize. Missing should remain missing. The missing-metrics guide explains how to preserve that boundary.
Do not create a fake portfolio exposure by multiplying hops
The ASML-Micron-Nvidia-Quanta path is useful because it shows an economic chain. Multiplying 7.64%, 17.62%, and 0.53% would not produce a valid portfolio exposure. Each percentage has a different company-level denominator.
Portfolio exposure requires actual holding weights plus an explicitly defined methodology. The portfolio concentration analysis shows how relationship overlap can be used without inventing a multi-hop percentage.
A practical diversification matrix
| Axis | Main question |
|---|---|
| Sector | Are the holdings classified in different industries? |
| Country | Are they exposed to different jurisdictions and macro environments? |
| Supplier | Do they rely on the same critical upstream companies? |
| Customer | Do they depend on the same large buyers? |
| End market | Are they tied to the same demand cycle? |
| Network | Do apparently different holdings converge on the same bottleneck or cluster? |
A portfolio can score well on one axis and poorly on another. That is the point.
Where the Altsets interfaces fit
The free workspace is useful for visually checking whether holdings converge on the same nodes. An LLM using MCP can turn that map into follow-up questions such as which shared relationships are quantified or which second-order paths deserve investigation. The API is more appropriate when the same overlap calculation needs to be run across a large portfolio or repeated historically.
The Altsets documentation covers the available interfaces. The underlying diversification logic should remain the same regardless of interface.
A repeatable supply-chain diversification review
- Start with the portfolio holdings.
- Record sector and country diversification normally.
- Map important suppliers and customers for each holding.
- Find shared direct counterparties.
- Trace only relevant second-order paths.
- Classify common end-market drivers.
- Distinguish structural overlap from quantified overlap.
- Do not multiply percentages across network hops.
- Use holding weights only inside an explicitly defined portfolio model.
- Compare the supply-chain result with the original sector and country view.
The hidden economic-cluster guide explains how to discover connected market communities. This article applies that idea to a different decision: whether a portfolio that looks diversified on paper is actually exposed to the same economic chain. For relationship definitions and limitations, read the Altsets supply-chain data methodology. Browse Supply-Chain Data Use Cases for deeper portfolio and company workflows.