A Portfolio Has Demand-Shock Risk and Supply-Shock Risk. They Are Not the Same.
September 14, 2026
Altsets
Research by Altsets Research
Customer dependencies create demand-shock exposure while supplier dependencies create cost, availability, and production exposure. Mapping them separately reveals why two portfolios with the same number of holdings can fail for completely different reasons.
Data used:Altsets Supply Chain Intelligence: 90k+ entities, 400k+ relationships, 20+ years of history.
Key findings
- The supplied Micron network contains both a major Nvidia customer relationship and a material ASML supplier relationship, illustrating why one holding can belong in both demand-shock and supply-shock stress tests.
- The supplied LG Energy Solution-Tesla and HPE-Microsoft relationships add separate external demand nodes, showing why dependency diversification should distinguish customer concentration from supplier concentration.
Demand-shock risk and supply-shock risk should be mapped separately across a portfolio. A holding can be vulnerable because an important customer stops buying, because an important supplier cannot deliver, or because both sides of the business are concentrated at once. Combining those channels into one generic "supply-chain risk" score hides the reason the position can fail.
Micron carries both demand-side and supply-side exposure
The supplied Altsets data shows Nvidia representing 17.62% of Micron revenue. That is a demand-side dependency. If Nvidia reduces purchasing or changes product demand, the first Micron risk is that an important source of revenue weakens.
The same supplied network shows ASML representing 11.91% of Micron's cost base, with a 3B USD relationship size. That is an upstream exposure. If the problem is equipment availability, pricing, or technology access, the risk enters through Micron's ability to produce rather than through customer demand.
Those two edges should not be collapsed into one number because the percentages have different denominators and the shocks affect different parts of the business.
LG Energy Solution is a clearer example of demand concentration
Tesla represents 19.03% of LG Energy Solution revenue in the supplied Altsets relationship, while LG Energy Solution represents 3.41% of Tesla's cost base. For an LG Energy Solution holder, Tesla belongs primarily in the demand-risk map. A deterioration in Tesla purchasing can affect supplier revenue even if LG Energy Solution's own production system is functioning normally.
That is a different portfolio risk from a company whose main vulnerability is a difficult-to-replace upstream input. Both positions can be "supply-chain exposed," but they should not be stressed with the same scenario.
HPE adds another external demand node without repeating the same customer
The supplied HPE customer set includes Microsoft at 561M USD, Swisscom at 203M USD, Home Depot at 109M USD, and Volkswagen at 64.5M USD. Microsoft is the largest of those four displayed relationships by a wide margin.
An investor holding HPE, Micron, and LG Energy Solution therefore owns three companies with materially different external demand nodes: Microsoft, Nvidia, and Tesla. That is more informative than simply counting three holdings. The investor can ask whether the portfolio's downside is clustered around the same demand source or spread across several unrelated customers.
The portfolio should be stressed by failure mode
A demand-shock stress test asks what happens when important customers buy less. A supply-shock stress test asks what happens when inputs become unavailable, delayed, or more expensive. Some holdings belong mainly in one test. Others, like Micron, can appear in both.
This separation produces cleaner portfolio decisions. A portfolio already heavy in customer concentration may benefit more from a new holding whose demand base is independent even if the new stock belongs to the same broad sector. A portfolio already concentrated around one critical supplier may need a different type of diversification.
The goal is not to produce one synthetic exposure percentage. The goal is to know which holdings fail for the same reason.
The conclusion is that dependency diversification has at least two axes
A portfolio can be diversified across customers while remaining concentrated across suppliers, or vice versa. The supplied Altsets relationships show why those channels need separate maps. Nvidia is a meaningful demand dependency for Micron, ASML is a meaningful upstream relationship for Micron, Tesla is a major customer for LG Energy Solution, and Microsoft is the largest of HPE's four displayed relationships. Those are different event paths even though all belong to one dependency network.
The portfolio concentration guide explains how repeated counterparties create hidden overlap. The customers-versus-suppliers guide explains why the side capable of breaking the thesis should determine where the investor focuses first.
For relationship definitions and evidence limits, read the Altsets methodology.