A short thread on an obvious selection effect with some big consequences.
The social networks that are huge and very powerful now are the ones that grew the fastest. All else equal, these tend to be those with compelling products, but also another crucial thing:
1/
Being willing to make most trade-offs in favor of growth during a crucial period, which often was pretty long.
That process isn't pretty: it involves being willing to manipulate users and operate as many viral loops as possible, as long as they don't have a *growth* downside
2/
There's also a large, and maybe more important, effect on corporate culture: the people who grow most powerful and influential at the company during this period are the ones who were willing to give up a lot of other things for growth.
3/
So now you have these companies that influence world events at the highest level. And, just by the nature of how they got to be that way, their senior leaders are selected on having been obsessed, to the exclusion of other values, with user growth.
4/
Not all the leaders, of course, but the directional effect is there.
And now, these are the people (naturally - who else?) who run earnest corporate responsibility initiatives to make the platforms serve society better.
5/
Facebook is, of course, the perfect example of this, but not the only one.
So part of what we are learning about in the early 21st century is what a politics and a culture presided over by growth hackers looks like.
6/
I think a recognition of this phenomenon is why some of the people who understand these companies (and the leaders involved) best feel darkest about things.
Thread inspired (most proximately) by this useful history
This terribly misguided paper is making the rounds.
This thread is to make it common knowledge what is wrong with it.
The basic thing: all modern economic theory allows for a gap between individual maximization and efficiency, whatever you mean exactly by each of these.
The first welfare theorem (individual optimization implies social efficiency) breaks down in the presence of frictions -
e.g., incomplete markets, asymmetric information, externalities, and market power.
Most economics today is about these frictions.
2/
Now, the paper has some halfhearted recognition of this, but says, effectively
"Well, you know, there is some meta-stage in which institutions are chosen, and economics assumes that this choice will be made to kill all frictions except the efficient ones."
3/
a few notes on it from an economist studying network theory
The striking thing about César's hit 2009 paper on economic complexity is that it doesn't mention eigen-anything and seems surprisingly disengaged from network theory.
The economic complexity index that Hidalgo and Hausman propose in "The building blocks of economic complexity" is a very close variant of Kleinberg's very famous 1999 HITS algorithm.
It's not clear whether they're aware of this connection, but in any case
2/
economists writing about networks in 2009, such as Jackson, Acemoglu, myself, and many others would have probably written the paper differently --
with a clearer consciousness to our big debt to the prior study of eigenthings as centrality measures!
I don't care at all about homework being done with AI since most of the grade is exams, so this takes out the "cheating" concern.
Students seem motivated to learn and understand, which makes the class very similar to before despite availability of an answer oracle.
2/
It's possible that (A) all the skills I'm trying to teach will be automated, not just the problem sets AND (B) nobody will need to know them and (C) nobody will want to know them.
Notice: A doesn't imply B and B doesn't imply C.
3/
A survey of what standard models of production and trade are missing, and how network theory can illuminate fragilities like the ones unfolding right now, where market expectations seem to fall off a cliff.