We have very small companies (in the low tens of millions) that make decent SoC. Nobody is going to make 4G anything for less than the tens of billions). The en

We have very small companies (in the low tens of millions) that make decent SoC. Nobody is going to make 4G anything for less than the tens of billions). The entire market cap of ARM is half that of Qualcomm (and around a third before they lost a quarter of their valuation). Building modems is way harder than building SoCs (all the SoC challenges plus all the modem ones too). It's taken a decade of work on 5G to just begin preliminary sampling (How many SoC generations would that be?). You also have to keep in mind that it's not just the modem in the phone. You have huge software stacks, protocol design and testing, and an entire cell tower worth of electronics to design and build. There's no free lunch. You have to strike a balance between forcing the cost onto the service provider (higher cell bills and/or slower adoption) and forcing the cost onto the phone. Considering that Qualcomm's net sales to R&D ratio is basically unmatched (most companies spend half what Qualcomm does by that metric), it's not easy to make the case for gouging in my view.