My doc2pdf will take .doc and .docx files, and using a couple programs including with OS X 10.8, makes PDF's from them.

Prior to 10.8 or maybe 10.7, the PDF conversion was done with a program found at /System/Library/Printers/Libraries/convert. Now it's done with /usr/sbin/cupsfilter and the options are a bit different.

It will:

• find the crappy doc(x) files and use textutil to make html files
• make sure they stand out with a .temp.html name.
• take those html files and 'print' them to PDF
• remove the those temp html files.