I'm a participant in a study that's exploring the genetics of intelligence. From twin studies we know it's at least partly genetic, and by sequencing the genomes of a large number of ideally-smarter-than-average [1] people we should be able to figure out how much and, more importantly, via what gene variants. Now maybe this problem will turn out to be too hard, with huge numbers of tiny changes interacting in an incredibly complex way, but what if it's not? What's the effect of being able go from a genetic test to a decent estimate of intelligence?

The most obvious application is people choosing to have smarter children. You could extract eggs, fertilize them, let them divide a few times to where they're big enough that they can afford to lose a cell but not so much that you have differentiation, extract a cell from each embryo, amplify the DNA of the extracted cell, test its genetics, and then only use the embryos that test highly. (Wikipedia: PGD) A big downside here is cost: at least initially it would be extremely expensive. When partially established it might be around $40k: it's a lot like having your egges frozen for future use, but with a few additional difficult steps, and that costs about $20k On the other hand it's much more broadly applicable than saving eggs for later, so it could have higher volumes, and so might get even cheaper. Still, it's basically always going to be thousands of dollars more expensive than just making a baby yourselves. It would probably be in the range of international adoption and IVF: expensive, but affordable for middle-class Americans [2] who want it enough.

One question is if this were available, would people actually use it? Via egg and sperm donation it's already possible to choose a child who is more likely to have all sorts of characteristics, including a greater chance of being really smart. Yet I haven't heard of a couple capable of conceiving a baby the traditional way who instead decided to choose their child's genetic parents. Among people who for reasons of infertility use donated genetic material, my understanding is that the intelligence and accomplishments of the donor are a consideration, but far from the only one. [3]

Let's say people do use it, though, or at least enough use it to start having an effect. It would become another of the many ways rich people can use to help their children be more successful people. Unlike many of those existing choices, however, it's mostly not a positional good. That is, if you donate lots of money to Harvard and your kid is admitted, a lot of the value to them comes from you managing to secure for them a scarce resource (a Harvard degree). Similarly if people use genetic testing to have their children be as tall as possible, that helps their child relative to other children but isn't otherwise useful. To some extent intelligence does work this way, where in competition for limited resources smarter people can have an advantage, but for the most part having a smarter population should on net be beneficial.

On balance I think this study is likely to be positive in its effect on the world, via it just being broadly useful to have more smart people, though it's always hard to predict the outcomes of new technology.

(Is this why I signed up? Nope! They said participants would get copies of their sequenced genomes and I was curious to see what's in mine. I sent in my spit a year and a half ago, however, and haven't gotten the results back yet.)

[1] They were selecting people that had one of: (a) PhD in the hard sciences from somewhere famous, (b) honorable mention on the Putnam, or (c) high SAT/ACT/GRE scores. I qualified via SAT scores, which was probably the easiest way in.

[2] But note that this is not the same thing as affordable in general: even an American at the poverty line is richer than 90% of people in the world. (And that's after adjusting for PPP.)

[3] I remember reading somewhere that attractive baby pictures had the biggest effect on likelihood of being chosen as a donor, but looking now I'm not able to find this. Or, in fact, much of any research on how people choose donors. (I did find a 1994 survey of college students, but it's kind of an unrepresentative mess.)