No I have no link.
And with seeing the discussion I see that it may well be that this is not necessarily time constant, but may change. A milieu study would be probably more stable, as income has such a strong age dependency, and e.g. strong gains in the student milieu may make the voters of a party poorer in average, but reflect a base of 'economic winners'.

Der Amerikaner ist die Orchidee unter den Menschen
Volker Pispers