This is interestingly linked to the Flynn effect (secular trend of rising intelligence) in some of the literature.
In the case of at least one of the
other trends, height, all four
features of the IQ paradox have been
documented, thus creating a “height
paradox” identical to that seen in
IQ. Even one of the earliest MZA [Monozygotic twins reared apart] twin
studies of height, conducted in a
U.S. population that was much shorter
than today’s population, showed the
reunited twins to be remarkably
similar in height (Newman, Freeman, &
Holzinger, 1937). This would sug-
gest that whatever presumed
environmental factor was stunting the
population must have been so uniform
in its effects that it did not matter
whether monozygotic twins were raised
in the same or different homes; they
were still stunted to virtually the
same enormous extent. Also, there is
no evidence of any birth order effect
in height (Ernst & Angst, 1983).
[...]
One major difference between research
into the height and IQ trends,
however, is that among researchers
investigating the height trend, there
is not the same general recognition
that the high heritability of the
trait poses theoretical problems for
environmental hypotheses as there
is among IQ researchers.1
This is all from Mingroni (2007). He argues that heterosis / hybrid vigour / outbreeding elevation (basically the opposite of incest) is (part of the) the reason. Basically people are more mobile and no longer have to sex up their cousins, etc. It's been a while since I read the article, but it's a good review. According to him, most of what he says about IQ applies to height as well. I think he sums the evidence up fairly and Psych Review is a high-profile journal. The author is pretty elusive though – can't find out much about him.
1 Mingroni, M. A. (2007). Resolving the IQ paradox: Heterosis as a cause of the
Flynn effect and other trends. Psychological Review, 114, 1104.