Sunday, January 13, 2013

Social connectivity and nationality

We're all in one big, highly connected social network, a la the six degrees of separation theory.

And in this network, nationality is a strong determinant of connectivity between two people.  If you are American, an American stranger is likely to be much "closer" to you than a non-American stranger, in the loose sense that there are more and shorter paths of social ties connecting you to the American stranger.  There's a denser web of connections.

And so it makes sense for an American to care about American strangers more than non-American strangers.  You care about your friends, and your friends care about their friends, so you indirectly care what happens to the friends of your friends, even if they aren't your friends.  And those friends have their own friends, and so forth.

This is, perhaps, a counterpoint to a previous post of mine.  It is also a counterpoint to a stance Steve Landsburg has frequently adopted some version of:
We need to care about others. We need to care about those who are close to us, and we need to care about strangers. But to care more about strangers who happen to be American than about strangers who happen to be Japanese or Mexican is an expression of the basest and most wrong-headed instincts that a person can have. 
I like where Landsburg is coming from.  But actually, the moment you allow yourself to care more about someone who is closer to you, the conclusion that you care more about American strangers follows quite naturally.  Because American strangers are closer to you than non-American strangers.  I'm not saying you couldn't tell a story that made equal treatment of all strangers best, but it isn't the default, and violating it isn't automatically wrong-headed.

We can try to rescue the sentiment by declaring you should care equally about two strangers who are equally connected to you, regardless of nationality.  But that is not particularly relevant when we are concerned with, say, the optimal immigration policy.  Because the strangers on either side of the border fence are not equally connected to you.

Speaking of immigration policy, I love this short Landsburg piece, in which he calculates a bound on just how much more we would need to care about ourselves to justify our policies:
How do you justify a border fence?...exactly how much are you willing to hurt a foreigner to help an American? Is a foreigner's well-being worth three-quarters as much as an American's, or half as much, or one-quarter as much? ... Let's do the math.
[Emphasis added]

Landsburg's first point is that there isn't actually a tradeoff between the wellbeing of Americans and immigrants.  In fact, many good things happen to us when we let in immigrants:
Virtually all economists agree that immigration makes us richer, not poorer. Every immigrant is a potential trading partner, a potential employee, and a potential customer.
But I have some bad news, in light of our social network analysis above.  To let someone into a country is to integrate them into the social network of that country.  Americans' wellbeing is affected by their social network, so naturally they will prefer some networks to others.  Indeed, Americans may choose to keep immigrants out because Americans don't want to care about those immigrants the way they care about other Americans

Relatively speaking, America doesn't have much of a poverty problem.  That means that Americans, who care most about Americans, don't have to feel too bad about poverty.  If we integrated an impoverished country into our own, it could actually make Americans feel worse.  And so Americans might choose not to integrate a bunch of worse-off people into their social network, even if it would lift everyone's wellbeing before social effects were accounted for.

If you ask me, there is too little immigration into this country.  But I can understand why people might feel differently.  Nobody cares about all people equally, and social connectivity clearly matters.  It is ugly to care about some people less because of an arbitrary label like nationality, but maybe that's not what's going on.  Maybe nationality is just a useful observable that's correlated with the one thing people truly and rightfully care about, namely social connectivity.

Of course there are plenty of lousy anti-immigration arguments.  But there is at least one good one.

1 comment:

  1. Interesting post. You could take the theory a step further, and use it to analyze heterogeneity in nationalist preferences. Some thoughts on this angle: WASPs are less likely to be closely connected to a potential immigrant stranger than first or second generation immigrants. Alternatively, you could look at demographic makeup of states to get state-by-state (or county-by-county) variation in connectedness to those on the other side of the fence.

    Surely, this analysis would have confounds, but it seems that the predictions of a social network model would suggest a particular pattern of attitudes toward immigrants that would distinguish it from classic economic arguments that condition predictions on social status, occupation, etc. b/c of how the economic gains/losses from immigration are distributed.

    ReplyDelete