Impossible to ReplicateWide moat

The durability of ASML's monopoly rests on the near-impossibility of replicating an EUV machine, and it is worth dwelling on just how deep that barrier runs. EUV took ASML more than twenty years and tens of billions of euros¹ to make work — solving physics problems many experts believed were insurmountable, such as generating usable 13.5-nanometer light by blasting molten tin droplets with a laser fifty thousand times a second, and steering that light with mirrors so perfect that, scaled to the size of a country, their imperfections would be measured in millimeters. This is not engineering that can be reverse-engineered from a finished machine; it is accumulated, hard-won mastery of the outer edge of physics.

Crucially, the knowledge is not codified in any transferable form. It is distributed across ASML's own physics and software, its unique supplier ecosystem, and the tacit experience of thousands of engineers who learned by doing, over decades of failures. A competitor with unlimited capital could not buy this, because it is not for sale; could not copy it, because it is not written down; and could not shortcut it, because the learning is sequential and slow. This is why the barrier is a moat rather than a mere lead: a lead can be caught by spending more, but a moat made of two decades of irreproducible, distributed know-how can only be crossed by spending two decades — during which ASML moves further ahead. It is the rarest kind of advantage: one that time makes wider, not narrower.