Trump Appoints China Hawk to Trade Council
 
From the Financial Times:
 

The day after Trump's election I wrote that the tech industry should hope that Peter Thiel was right that Trump should be taken seriously, but not necessarily literally (the seriously/literally formulation actually originated in this Atlantic article):
 

However, Trump's appointments to date certainly suggest that he should be taken seriously and literally, and the Navarro appointment is a perfect example; to say he is a China hawk is an understatement. And, in terms of tech, Navarro has taken aim at Apple explicitly. Here he is in The Huffington Post in 2012:
 

 
What makes Navarro's critique challenging is that it's not wholly wrong, at least from the American worker perspective, yet it's not particularly actionable. The most obvious example is currency manipulation: there is broad agreement that China did keep its currency artificially low for many years, but in fact over the last couple of years China has been propping up its currency in the face of significant capital outflows. The reason this is instructive is that it is 2016: reasonable people can debate about the effects of past Chinese currency manipulation narrowly and previous U.S. trade policy generally, but the only thing that can be changed is the future. What good is accusing China of devaluing their currency when the country is currently doing the opposite?
 
Apple and Chinese Manufacturing
 
This applies to Navarro's other complaints, and for at least two of them the Apple example is indeed quite useful. Consider this New York Times piece from 2012, which is built around the story of how the original iPhone switched from a plastic screen to one made of glass in the months between the products January introduction and June launch. A central piece of the story are illegal (under the WTO) subsidies:
 

The problem now is that not only do those glass-cutting plants exist, but so do a whole host of other plants assembling all the various components of an iPhone; yes they may have been built with illegal subsidies, but the very fact they exist gives them a massive pricing advantage because their costs are depreciated, plus the logistical advantages from all being in the same place.
 
This bleeds into the worker question: first, as much as Apple may protest to the contrary, Chinese workers are cheaper (although the cost difference is significantly less than it was a decade ago); estimates vary as to how much more it would cost to assemble the iPhone in the U.S. presuming Apple used the same methods, but 5~10x seems to be in the ballpark (i.e. from about $6 per iPhone to $30~$60).
 
It's not just cost, though, as the iPhone screen story illustrates; it's flexibility:
 

Foxconn disputes the "middle-of-the-night" narrative, but the speed with which the manufacturer can redeploy massive number of capable workers is without question; whether or not this is "mistreatment of workers" depends on your frame of reference. Chinese factory workers do endure conditions that most Americans wouldn't; those conditions — particularly at Foxconn — are also far better than the hardscrabble rural existence most of those workers are fleeing from, and a major factor in lifting China out of poverty, and by extension, reducing global poverty to record lows.
 
Then there is the third factor, expertise. Again from that story:
 

This is certainly an education issue — the sort of industrial engineers described here are the products of vocational-type schools, more skilled than high school graduates but not as much education as university student, which is why Tim Cook raised this issue with Trump — as well as sheer volume and cost considerations. But it's also a problem that has a self-fulfilling nature to it: the fact that these jobs are now in China means that China is preparing industrial engineers to fulfill them — and, by extension, the U.S. is not. And, more optimistically, many of the people who might have done these jobs are now employed elsewhere, resulting in a net gain for the U.S. economy. Regardless, the problem is similar to the currency manipulation issue: whatever may have happened in the past can be debated about but the decisions about the future start from the reality of the present.
 
Automation and Apple's Risk
The opening anedcote in that story is Steve Jobs' declaration to President Barack Obama that manufacturing jobs weren't coming back; in fact, though, Apple with much fanfare did start manufacturing the Mac Pro in America in 2013. This week Mark Gurman wrote in Bloomberg that it is a move Apple may have come to regret:
 

Forgive me while I venture into product territory, but I've thought from the beginning that the current Mac Pro was a mistake: it seems to me that a product focused on the most demanding users should have more flexibility, not less. That applies both to extensibility (the Mac Pro can only be extended via Thunderbolt, not internals) but also to use case: the Mac Pro was built around high end video cards, which not only meant users who needed a high end computer but not said video cards (which, it should be noted, are about more than graphics these days) were penalized, but also that the Mac Pro was committing itself to an even more aggressive update cycle given that video cards are progressing much more quickly than general purpose processors.
 
Moving manufacturing to the United States, though, was only possible from Apple's perspective because of automation, and here it is important to understand the economics of said automation. Human labor assembly is a marginal cost: every additional unit produced requires X amount of human labor, which must be paid for. Robots, though, are a fixed cost: you pay for the robots up-front, but the payoff is that the assembly of an additional unit has zero marginal costs (excluding things like electricity, maintenance, etc.). The implication of this shift from marginal to fixed costs is that there is a heavy incentive to stick with a specific design: the more units you produce the more leverage you get from those fixed costs; on the flipside, any change requires significant capital investment to update the robots. In the case of the Mac Pro, the question I have is to what degree does Apple find itself handcuffed by its investment in automation, particularly if it wants to change the form factor?
 
What China offers, for all of the reasons listed above, is the chance for the company to have its cake and eat it too: low (marginal) assembly costs plus high flexibility; manufacturing in the United States would mean either higher assembly costs — if Apple can even hire enough workers — or decreased flexibility thanks to the constraints of automation.
 
This is why Navarro's appointment specifically and Trump's clear signaling about China generally should greatly concern Apple in particular: any change from the status quo, whether that be shifting manufacturing to the United States or some sort of tariff on imported goods, will result in some combination of reduced flexibility, lower margins, and higher prices. The kicker, though, is that Apple would pay on both ends: China has already singled out the iPhone as a means of retaliation, which means any sort of trade war will hurt Apple in its two biggest markets simultaneously.
 
That said, it should be noted that Apple doesn't exactly have karma on its side: the fact that iPhones are manufactured outside of the U.S. is a critical component of their tax avoidance efforts. The company is right to argue that it is a failure of measurement that the iPhone alone contributes billions to the U.S.'s trade deficit when the vast majority of the value is captured by the United States (and other countries like Japan and Taiwan; China's value capture is quite small); their righteousness — and insistence that manufacturing in the U.S. is impossible — is diminished by the fact they don't pay taxes on a significant part of it.

Vladimir Poutine a ordonné, jeudi 22 décembre, le renforcement en 2017 de la force de frappe nucléaire russe. Il souhaite la rendre capable de percer tout bouclier antimissile, comme celui que Washington entend déployer en Europe orientale.