Matt's Thoughts In Between - Issue #38

Forwarded this email? Subscribe here.

There’s rightly been a lot of concern in the UK and US about wage stagnation, inequality and falling social mobility (and their potential populist political consequences). So Russ Robert’s new post, which claims that most studies on the topic are deeply misleading and that in fact most of the gains from economic growth go to the poorest, is both contrarian and fascinating.

Roberts’ key argument is that studies such as Piketty and Saez’s make the crucial error of comparing the group of people who happen to be in the bottom (or top) quintile at one point in time with the group of people who happen to be in that quintile some time later. But in fact, argues Roberts, these people are not the same people: instead we follow the individuals from, say, the bottom quartile at one point in time and ask how they are doing some time later, all the traditional conclusions are reversed. 

Roberts looks at panel data that follows individuals from 1987 to 2007 and finds, among other things:

This doesn’t invalidate many important narratives (e.g. relative social mobility is still very low), but it repays careful thought. If we want a fairer and more equal world, we need to start with an accurate analysis - and Roberts’ post suggests it’s more complicated than it first appears. 

We’ve talked before in TiB about China’s bid to build a domestic semiconductor industry, partly for national security reasons. Nikkei (who had a great piece on the topic I linked to back in May) have published an interesting update this week. 

It identifies a vicious cycle that I expect we’ll see perpetuated more and more: Chinese companies seek joint ventures with US and other technology leaders; these leaders fear IP theft and technology transfer; interest in JVs diminishes on the US side; and so the Chinese government provides more state funding to semiconductor and sensor manufacturers, thus deepening the national security/arms race dynamics.

It’s interesting to think about this in the context of this story, via Tyler Cowen, about a Canadian firm installing Chinese-made underwater monitoring devices by a US submarine base. Just a few years ago this might have seemed an advert for the interconnected nature of the global economy. Today, as Cowen notes, it looks more like a national security lapse that will seem extraordinary in retrospect. 

Lambda School, the coding boot camp whose model we’ve discussed before, announced recently that they’re expanding to Europe, fresh off the back of funding round led by Google Ventures. The extraordinary income increases Lambda reports suggests it’s sure to be very popular. 

This got me thinking, though, about Dan Wang’s (see previous TiB coverage) excellent essay from last year asking, “Why do so few people major in computer science?”. It’s worth reading the whole thing, but in short Wang concludes that it’s quite hard to explain. Computer Science (CS) is both highly paid and increasingly prestigious, but is far less popular than subjects like mathematics and physics. 

One possibility that Wang discusses is that perhaps the relative scarcity of CS degrees is a thing of the past and by 2015 CS was on a steeper growth trajectory. Wang considers the idea that the film the Social Network nudged a lot of people into startups and, by extension, CS. And perhaps this changed the culture of CS forever. I enjoyed this line from Wang, which may explain tech’s fairly recent(?) “bro-ification“:

Getting from here to there. Beautiful thread of “isochronic maps”

In Jeff we trust. What’s the one institution trusted by both Democrats and Republicans?

Hard bargains. What’s it like to negotiate with Bill Gates and Steve Jobs?

Funnier than fiction. Perhaps the most amusing story ever told about Google.

Are they ordering hockey sticks? The rise and rise of package volume in China

I love getting your thoughts and feedback. If you enjoy TiB, please tell someone else who might like it. You can just forward this email to them or send them here.

Until next week,

Matt