Your Tier 1 is managing
the allocation. You just
don't know it.

Ford and GM have signed direct memory and storage supply agreements with Micron, bypassing the Tier 1 module suppliers who traditionally sit between OEMs and semiconductor manufacturers. On the surface, this looks like a post-2021 lesson applied: go upstream, lock in supply, reduce exposure to intermediary fragility. That framing is not wrong. It is just incomplete in a way that matters for anyone outside automotive.

The more consequential part of this story is what the move reveals about the information architecture of the existing supply chain. When Ford and GM needed to understand their actual position in Micron's allocation queue, they could not get a clear answer through their Tier 1 suppliers. The Tier 1 was the intermediary for physical product and for information. That opacity is not a bug in the relationship; it is often a feature from the Tier 1's perspective. Managing allocation visibility is leverage. Knowing that your customer does not know their true queue position means you control the conversation during a shortage.

Sub-tier opacity is not specific to semiconductors. It exists wherever a critical input is allocated by a concentrated supplier base, flows through an intermediary, and is not tracked directly by the buyer. Electronics, industrial equipment, specialty chemicals, defense components: the pattern repeats. What is new is that Ford and GM have formalized the act of breaking it, at scale, in public. Precedents in procurement travel faster than most people expect.

Once an OEM establishes a direct relationship with a Tier 2 or Tier 3 supplier, the commercial conversation shifts permanently. Micron now has a direct line to Ford's and GM's engineering and procurement teams. That changes how specifications get written, how roadmaps get shared, and who has pricing power in the next contract cycle. The Tier 1 module supplier does not disappear, but their role compresses: assembler and logistics node, not strategic intermediary. If you are a procurement leader whose category has a similar structure, the question is not whether this will happen to your supply chain. It is whether you want to be the one who initiates it, or the one who reacts after the next shortage.

Samsung, SK Hynix and Micron control roughly 90% of global DRAM production (Omdia, Q1 2026). Within automotive memory specifically, Micron holds the largest single share (TechInsights, 2024); Micron is the only one of the three with U.S. manufacturing capacity (Manassas, VA).

DRAM prices for automotive-grade LPDDR4 are up roughly 70% year-over-year since December 2025, per S&P Global Mobility, with older chip generations expected to keep rising through 2026 as supply for legacy nodes tightens further. For an OEM with a multi-billion-dollar annual semiconductor spend, that number is already a line item in every CFO conversation. The financial exposure for procurement teams in other sectors is less visible, which makes it more dangerous. If your category has a critical input flowing through a single Tier 1 intermediary, and that input is allocated by a concentrated supplier base, you are carrying a contingent liability that does not appear on your balance sheet until a shortage forces it into the open. The 2021 semiconductor crisis removed an estimated $210 billion in automotive revenue in a single year, according to AlixPartners. Not because chips did not exist. Because they were misallocated, managed by intermediaries whose priorities did not align with their customers'.

The finance translation for non-automotive teams: if your inventory base for a critical input category is in the range of €5M–€20M, and that category is allocated by a supplier your Tier 1 manages on your behalf, you should be asking what it would cost to buy queue position directly. The Ford and GM model suggests the answer is a premium, paid explicitly, in exchange for named allocation. That premium is almost certainly cheaper than the working capital destruction and lost revenue from a three-month production gap. The question for your CFO is not whether to pay for sub-tier visibility. It is whether you are already paying for it indirectly, through your Tier 1's margin, without getting the protection in return.

I do not sit inside procurement myself, but from where inventory strategy intersects with it, I keep seeing the same pattern: suppliers get treated as a third party to manage, not as an actor inside the process. That is exactly where priority gets lost, quietly, long before a shortage makes it visible. And it is not only about price. Paying on time, or negotiating a strong discount, does not automatically make you a preferred customer in your supplier's internal ranking. The relationship runs on more than the payment terms written into the contract.

What Ford and GM have done is refuse to accept that information asymmetry as a cost of doing business. They made it structural: a direct agreement that forces allocation visibility into the contract. Why do more procurement leaders not do this? In most companies, the Tier 1 relationship is treated as a boundary: something you manage, not something you look through. That boundary is comfortable. It means fewer suppliers to manage, cleaner contracts, a single point of contact. The discomfort only arrives when supply tightens and you discover that your Tier 1's "manageable" was their assessment of their own position, not yours.

In your top five critical input categories, do you know your actual queue position with the Tier 2 supplier? Not what your Tier 1 tells you. What the Tier 2 would say if you called them directly tomorrow. Most teams do not have that answer, and I do not think it is usually laziness. Mapping a supply chain three or four tiers down is genuinely hard: a Tier 3 or Tier 4 supplier in a place like Vietnam may not have a website, let alone a disclosed customer list. The opacity is often structural, not negligent. What I would push back on is stopping there. For your top suppliers, the ones actually carrying allocation risk, a joint Tier 2 and Tier 3 mapping exercise done together with them, not despite them, is a reasonable ask. That is a different problem from full-chain visibility, and only one of them gets solved by a better contract.