Lab Lemming asked one of those questions that one thinks is a complete no-brainer. Until people start chiming in with their opinions and you realize yet again that not everybody has the same take on the unspoken issues.
What institutional affiliation does one list on academic papers?
Lab Lemming posed a rather simple (and easily generalizable if you consider trainees) scenario:
If a private sector researcher is submitting a manuscript to an academic journal, and said researcher is concerned that the timescale of peer review might be longer than the timescale of employment (or even employer solvency) due to the current economic situation, is it considered unprofessional to use non-work contact information?
At the simplest level of the journal being able to contact this author, this is a no brainer. Certainly one should feel free to include the gmail or yahoo email address if one is concerned about losing access to an employer address. Of course this is mostly moot because for most modern journals with online submission, one can presumably change the personal contact information and instantly the correspondence will be appropriately routed.
In my view, the more interesting question here has to do with what goes under the author name as the local university, institute or company at which the author was employed when the study was conducted. Oops, just in writing this I'm showing my bias. I think that the affiliation is for the place where the author in question was employed when the (bulk of) the work was completed. So this does not change just because the company has gone belly up or the author has been fired or recruited elsewhere.
It is okay to have multiple affiliations listed if substantial work was conducted across job postings. If, however, a paper has been submitted, an author has moved and no additional substantial work has been done at the new posting then the appropriate thing to do is to list the "current address" in the Acknowledgment or special footnote (depending on journal practice). What one should not do is to change the affiliation to one's new employer as if that is the place at which the work in question was conducted.
Although I am not certain, it seems perhaps some of the comments at LLL may not see things this way.