Except that compounds with -er aren't that recent a development. Peacemaker, for example, first occurs only a few decades after pinchpenny and well before makepeace or pickpocket.
Simple agent nouns in -er go back to Old English and compound occupational terms (e.g. bellmaker, bookbinder, etc.) are common by 1400. Compounds without -er (e.g. wainwright, shepherd, etc.) are even older, with attestations before 1000. (These may appear to be N-V, but they're actually N-N; before the adoption of -er from Latin, the Germanic method of deriving agent nouns was to make weak nouns from verb stems.)
By contrast, I can't find an example of one of these V-N compounds from earlier than 1362 (cutpurse); I don't remember ever seeing one in Old English. It seems to have a been always a less common means of derivation that existed side-by-side with N-V-er for a while and has now almost fallen out of use.
no subject
Simple agent nouns in -er go back to Old English and compound occupational terms (e.g. bellmaker, bookbinder, etc.) are common by 1400. Compounds without -er (e.g. wainwright, shepherd, etc.) are even older, with attestations before 1000. (These may appear to be N-V, but they're actually N-N; before the adoption of -er from Latin, the Germanic method of deriving agent nouns was to make weak nouns from verb stems.)
By contrast, I can't find an example of one of these V-N compounds from earlier than 1362 (cutpurse); I don't remember ever seeing one in Old English. It seems to have a been always a less common means of derivation that existed side-by-side with N-V-er for a while and has now almost fallen out of use.