Recession may not mean more OSS
Strangely enough, I have seen more OSS-based projects during boom times than recessions. From past experience, what happens in a recession is the boards get even more risk-adverse. Whilst stretching existing kit and software to the limit becomes the order of the day, when it comes to new projects that can't be delayed or scrapped the drive is to get them in on-time and to budget ASAP, and so they see proven stacks as providing the best guarantee of that. Unless the open source solution is just as proven as the OTS (off-the-shelf) one, those whose jobs depend on it being right will plump for the OTS choice as they see it as being less of a risk.
I can remember going to the boardroom in another company in 2001 with two solutions for a business requirement - one of proven software (including Oracle) on UNIX servers; the other Red Hat on x86 with OSS software from several sources, which had not been tested before, and would need some inhouse coding to get it all to work. Despite the OSS solution being less than a third of the price (even after including a 200% costing for the inhouse work), we could only project the likely success of the OSS solution as a 90% certainty, whereas the OTS solution was already proven. The board went for the OTS solution.
I see it as more likely that existing kit will need to last even longer, so lots of upgrades and extended contract negotiations to look forward to. OSS like Linux will get a look in for situations when it can show it has a proven option with minimal disruption and risk.