Published Last Week

Howmet & Precision Castparts: Pricing Power & Negotiating with Rolls-Royce

Last week, Howmet declared a force majeure event, effectively claiming it will avoid its obligations if its cost of goods is hit by Trump’s tariffs. Whether the engine OEMs such as CFM and Rolls Royce will accept the force majeure is unclear. But this event hits on a specific topic we’ve been recently exploring: Howmet and Precision Castparts pricing power. 

This week's interviews with a Former EVP of Procurement at Rolls-Royce explores how the OEM sources and prices structural and airfoil investment castings, the core and most expensive parts in a jet engine. 

When there is new engine platform, OEMs play with the market share between Howmet and PCC:

The only real way through that with PCC and Howmet was continually playing with the market share. It was fortunate for me and others during that period because we brought the A350 volumes to the table. We introduced the new design Trent 1000 components, particularly turbines, two brand new business jet engines, and the Trent 7000 engine, which succeeded the Trent 700. So, with five engine platforms, when you sit down with these companies, they have leverage, but they also want a piece of the action. They're paranoid to the point of obsessiveness that the other guy might beat them with market share, so the conversations tend to be relatively easy. - Former EVP of Procurement at Rolls-Royce

But this is only when there are new engine platforms. Without new volume to sell, OEMs struggle to combat supplier pricing power:

Regarding pricing and leverage, which you mentioned a few minutes ago, you have no real leverage with PCC and Howmet, except for volume and engine platforms…the reality is that you've engineered yourself into single-source situations with two fairly aggressive large American corporations, and it's a constant struggle. - Former EVP of Procurement at Rolls-Royce

Supplier switching costs are high due to engine recertification costs:

Switching isn't as easy as you might think. It's not impossible, but for those types of components, the cost of changes can be prohibitive. You might need engine testing, engine recertification, and possibly even aircraft recertification. If you're going to switch, you would do so with new engines and new components - Former EVP of Procurement at Rolls-Royce

The interview also explores how Rolls sources parts from Asia, risks within the supply chain, and the shift between making and buying parts across the XWB platform. These interviews can be read alongside other research in our library:

McKesson Corporation: Medline Med-Surg Distribution Economics

This interview forms part of our coverage on Amazon's expansion into healthcare B2B distribution. A Former Medline VP explores how how med-surg distributors such as McKesson and Medline lose money on distribution:

An interesting benchmark for you, is that Medline and all the big distributors lose money on just the distribution. Distribution costs, depending on complexity, such as low unit measure distribution and logistics concerns, are around 7%. This includes warehousing and trucking, costing Medline or Cardinal about 7% of revenue. However, the distribution rates charged to hospitals are often below 1%, resulting in a negative 6% on distribution alone. While there is some back-end funding, like Johnson & Johnson paying two points below the line for sutures, you still can't beat the math—you're still underwater - Former VP at Medline

In order to break-even, Medline would need 30% private label penetration:

So why would any company do that? Because controlling the channel allows you to profit from your own self-manufactured or private label products. A distribution deal for Medline typically breaks even around a 30% Medline brand ratio. There are factors like margin and what you're selling on exam gloves and surgical packs, but as a rule of thumb, 30% is the breakeven point. At Medline, if it's less than 30% Medline brand, you're not making money on the deal, and it could be a loss. After 30%, it starts to be profitable, and at about 40% Medline brand, you have a really good, mature, profitable account for the long term - Former VP at Medline

This interview can be read alongside the following:

Amazon Robotics & Challenges with Humanoid Robots

A Former Amazon Robotics executive on why she believes humanoids may not be as effective as expected:

Humanoid robotics work is not just about making any robot function once. For business purposes, you want a robot that offers reliability, repeatability, and a clear understanding of its success and failure points. You should choose your robot based on these factors, along with cost. In all these aspects, humanoid robots fall short. Due to their design, humanoid robots inherently require an order of magnitude more motors to manage their functions compared to the load they need to lift. Moving an object from point A to B becomes a whole-body problem. You must consider the positioning of the legs, how to brace them to reach out with a hand, and maintain balance. By simplifying and removing the legs, you create a cheaper, faster, and more reliable robot. You improve every aspect of the solution by transitioning from a humanoid to a humanoid top on a mobile base. The only limitation is that it can't walk up steps, but how many places actually require robots to navigate stairs? In my opinion, humanoid robots address a problem that doesn't exist in most environments. I am concerned about the number of companies developing humanoid robots. There might be a few applications, such as disaster zones, but ruggedized vehicles might be more effective.- Former Amazon Robotics Executive

FormFactor vs Technoprobe: Intel's Customer Perspective

This interview with a Former Director of Assembly and Testing at Intel walks through how the company purchases probe cards between FormFactor and Technoprobe. Probe cards are technically mission critical and Intel dual-sources:

Generally, Intel wants to have two suppliers. We have what we call Multiple Qualified Suppliers (MQS) for critical components. Probe cards are considered critical components, so you want multiple qualified suppliers. You pick two suppliers, and the allocation of business is managed by the factory teams. They figure out how to allocate the business because it's not always 50/50. If one supplier has a better cost structure, they might get more allocation. If one supplier's technology performs better, they might get a higher allocation. If one supplier has better customer service, responsiveness, and field service engineers, they might receive a slightly higher allocation. - Former Director of Assembly at Intel

But mission criticality doesn't necessarily translate into pricing power:

In our case, there was significant pressure from our side, insisting that prices couldn't be raised. If a supplier raised their price, it would affect the 70/30 allocation, meaning they would receive a lower allocation. Part of the pricing strategy involved using a multiple qualified supplier strategy to push prices down or at least keep them flat, preventing any increase. Being in a close relationship with these suppliers, we know what their manufacturing process looks like. We have engineers on the floor all the time, observing everything. It's not hard to put together a should-cost analysis and say, "Based on their manufacturing process, this is their should cost." If you add a 30% margin on top, this is what we should pay, and not a penny more. That's how the negotiations went. - Former Director of Assembly at Intel

The interview goes on to explore why suppliers may lack pricing power and how OEMs negotiate probe card pricing and volume.

Judges Scientific PLC: Geotek & Multi Core Scanning Industry

In 2022, Judges acquired Geotek, its largest ever acquisition, which now accounts for ~30% of EBITA. This interview with a Former Geotek senior executive explores how the product technology works and Geotek's positioning vs competitors:

Geotek has shifted from instruments to services. They are still unique because they offer both instrument sales and services. This business model allows them access to a broader pool of customers. Other core scanning companies I've worked with in the past won't sell anything, which often results in no sale. Geotek's ability to offer both gives customers better flexibility. This is particularly evident in the oil and gas market, where services are rare, and sales are more common. In research and academia, it's always instrument sales, and in mining, there's a bias towards services, but instrument sales do happen. Geotek's flexible business model provides better market access. Their technology portfolio is still the broadest but is becoming less unique as other market players emerge, especially in mining. Notable companies like GeologicAI and Veracio are entering this space. - Former Director at Geotek, Judges Scientific

On coring demand:

Coring is even more unpredictable because it's government-funded. Most coring revolves around gas hydrate analysis, which are frozen methane accumulations in the subsurface. These are unconventional resources that governments fund to explore their potential for energy security. They're trying to move these from deposits into resources or reserves. Industrial partners see it as too risky, so governments invest instead. Only a few governments have the money or motivation to do this, mainly Japan, China, India, the US, and occasionally Norway - Former Director at Geotek, Judges Scientific

The interview goes on to explore how Geotek has grown the services business to dampen coring volatility and the underlying competitive dynamics for Geotek. This interview can be read alongside prior work on JDG: