When Managing the Supplier Is No Longer Enough
GE Aerospace has agreed to pay $11.75 billion for Consolidated Precision Products, a castings company most people outside aerospace have never heard of, and one GE has been buying from for more than 15 years.
CPP makes highly engineered castings that go into jet engines, and castings have become a persistent constraint on engine output. GE says it wants more capacity, but it also wants design and manufacturing closer together so it can apply its own operating system to the plants and deploy enhanced airfoil technology faster.
This is as much about control as it is about volume.
Walmart is a useful contrast. It built one of the world’s largest supply chains largely by getting exceptionally good at managing external suppliers and the information that flows between them, rather than owning every factory.
That is another answer to a critical dependency: get very good at managing the system around the supplier.
But there is a limit to what process management can solve.
Make-or-buy usually starts with cost. Can a supplier make this cheaper than we can, and can buying avoid tying up our capital? That logic holds until the dependency itself becomes strategic.
When a capability is difficult and slow to reproduce and determines whether your product ships, unit cost is only one part of the economics.
Picture the supplier risk register on your own program. There’s a tab called “single source” that nobody has opened since the last audit, and beside the supplier that really matters the mitigation column says “alternate vendor identified.”
Ask who the alternate is and you’ll often find a company that has never made the part, which is not the same as having a supplier.
If this supplier disappeared tomorrow, how long would it take us to recreate the capability?
Naming a company that could make something is different from having the tooling, people, process knowledge and qualified capacity to produce it.
Some inputs are easy to replace. Others are constrained but can be substituted with enough time and investment. Then there’s a smaller set of capabilities so difficult to reproduce that losing access changes whether you can deliver at all.
Ownership is one answer for that last group. Long-term capacity commitments or co-investment can reduce the risk without an acquisition.
GE says it will keep relying on external partners, but $11.75 billion is a strong signal that at some point managing a dependency is no longer enough.
Sometimes the obscure part buried deep in the supply chain is the thing the whole strategy depends on.
Which input on your own program would you have to recreate rather than replace, and does anyone above you know that?
Comments