The fact that you don't consider interoperability to be an "objective criteria of the merits of software" is a perfect example of someone inventing criteria that isn't objective.
Companies don't operate in their own private bubble. A person who defines bidding criteria for office software and who doesn't consider the effects of interoperability with suppliers and customers is most likely someone with an agenda.
In the case of Munich, it obviously wasn't a money saving agenda, since you said they rejected the low bid. I'd put my money on an anti-American agenda.
In the US, the low bidder is almost never rejected because of fear of lawsuits and corruption charges. Europe evidently has no such fear of screwing American companies. And why should they? They have so many vocal American supporters cheering them on... even here on Free Republic.
If you need that level of interoperability, then it would be in your criteria. If you don't, then it won't. It's that simple. But most organizations and companies do not have that much of a requirement for interoperability of document format. Basically, as long as you can read Word or Excel every once in a while, you're okay. I've worked in these places, from major universities to some of the biggest companies in the U.S.
Europe evidently has no such fear of screwing American companies.
You haven't been reading. Linux was ahead in the beginning because it came out ahead on a contracted third-party merit study. That Microsoft's price was so high was a definite nail in its coffin, but even lowering their initial bid so much was apparently not enough to overcome the shortcomings in other areas.