There are some benchmarks in the following presentation: http://www.slideshare.net/8kdata/toro-db-pgconfeu2014 where ToroDB was presented (PostgreSQL Europe Conference, Madrid, 2014, 2014.pgconf.eu).

I'd be very interested to know more about this too. There is considerable overhead for each row in PostgreSQL (24 byte header, which must be aligned). Obviously you'll save a bit if there are repeated keys higher up, but you'll pay for the foreign key references to those rows in the children.

It is true that PostgreSQL has such a high row overhead. But the storage savings are significant, in any way. Please see the presentation pointed out earlier.