Can you name a single occurrence where a publicly funded university in the US or Europe forced a researcher to retract a paper because it conflicted with the universities' commercial interests?
From my time at academia I also don't recall the university or research institute asking to check a manuscript before submitting it to a journal, I'd even go as far as saying that this was unheard of as it would go against the spirit of research.
And in terms of research outcomes the funding bodies usually want publications in high-impact journals, public acknowledgment or results that can be commercialized. There are disciplines where commercial funding plays a larger role (e.g. pharma, medicine), but for most publicly founded research it's a rather bold assumption to claim that the funding is conditioned on achieving a particular result and not on the general impact of the results.
Okay, so what do you call the hiring process at university if not selecting for what kind of papers will be submitted? Mostly selections for good reasons, sometimes less so (nepotism and indeed politics).
Besides, your basic point is not valid: retractions by employer are forcibly demanded commonly in cases of, first, plagiarism, fraud, error and in many cases there are political undertones of this. Granted, in the vast majority of cases it is the researcher using shady research practices because of the researcher's strong political persuasion, but the reverse happens too: universities with (large departments with) strong political opinions that punish researchers or force retractions.
The most famous of these, in my opinion, is the one study claiming vaccines cause autism. Of course, everybody who has family on Facebook is still angry this particular researcher for good reason, and he was definitely grinding a political axe, and well paid to do so:
Second, I would argue that a lot of fields of study have "existential problems" that cannot be discussed in papers, or too much in public at all, without consequences for the researchers. What I mean is that there are a bunch of disciplines that have particular vulnerabilities in their foundations, which boils down to famous, extremely authoritative papers that should have been retracted a very long time ago, but weren't. Of course there's easy targets here, such as the Taylor papers [1]. But it goes pretty far: the original autism research was found to have one purpose only: to justify killing children to purify aryan blood lines (research happened right before WW2) [2]. These essentially invalidate a (large part of) a field of study, expose it as a fraud, but they are too enmeshed and refuse to die despite being exposed (and of course, you could more benignly argue they have since expanded and while they don't reject exposed fraudsters like Taylor and -much worse- Asperger, or you could argue they help people, unfortunately for autism treatments research disagrees with that). But you will find them equally in more "pure" research subjects: 99.9% (sorry) of statistics applies only in number series where the law of large numbers applies (which boils down to meaning there is no change in the underlying phenomenon being measured). There's 2 problems with that: first, upon careful inspection it never does, everything is changing. Second, there is an extremely common case that is always excluded: you can measure a "proxy" number for a phenomenon (say, number of theft convictions indicates something about crime rate), but as soon as you make any modification of the source system you are not allowed to use the same measure again, because this act invalidates the law of large numbers by explicitly changing the underlying phenomenon. The problem is that this is a valid criticism of 99% of applied statistics, and we have no resolution to this issue. You could argue that applied statistics works and has positive results, and I would agree. But ... it doesn't make it right.
There are entire fields that are coopted by political attitudes: the medical study of drugs is one example. The relationship between violence and imagining/experiencing violence (famously through TV and/or video games). Just search for long term effects of LSD on Google scholar and you will realise there is more going on than mere mistakes, and it is not really different, but much harder to tell, for most psychiatric drugs, even Valium and Methylphenidate. Many papers that are obviously bullshit are not retracted and some papers that definitely aren't bullshit are forcibly retracted because of the position they defend (such as, for example, that the average use of any given psychoactive medication is fraudulent, and exceed dosage and duration limitations of the medicines, with full knowledge of a psychiatrist who doesn't see a good alternative, which results in extreme damage to patients).
Third there are results that entire field feel they just MUST find, except they never do: the big example is the (direct) physical changes resulting from various psychiatric conditions. Every single study, for decades, has been shown to have been falsified, with a bunch of retractions, but also plenty of unretracted but obviously bullshit research, and several forced retractions of research pointing out the fraud. Just to be clear: this is not talking about depressed patients, say, cutting themselves, and the cuts resulting in physical change, this is about direct physical change (usually in the brain) resulting from (usually) depression. This results in people retracting papers without consequences because all the field sort-of-kind-of-not-really-but-... agrees with the fraud. Again you will find all 3: retracted papers, retracted-but-we're-still-teaching-it-and-rewarding-the-researcher-for-political-reasons papers, and forcibly retracted despite being true papers.
"Matthew Stewart points out what Taylor’s enemies and even some of his colleagues pointed out, nearly a century ago: Taylor fudged his data, lied to his clients, and inflated the record of his success."
"The Austrian paediatrician Hans Asperger has long been recognized as a pioneer in the study of autism. He was even seen as a hero, saving children with the condition from the Nazi killing programme by emphasizing their intelligence. However, it is now indisputable that Asperger collaborated in the murder of children with disabilities under the Third Reich."
In what sense is retracting a deeply flawed ("fraudulent") study that was invented for a court case the same as a university pushing a researcher to not publish or retract a paper because it paints the university in a bad light? It seems like you're trying to paint all retractions as equivalently "political", but that's obviously not the case.
From my time at academia I also don't recall the university or research institute asking to check a manuscript before submitting it to a journal, I'd even go as far as saying that this was unheard of as it would go against the spirit of research.
And in terms of research outcomes the funding bodies usually want publications in high-impact journals, public acknowledgment or results that can be commercialized. There are disciplines where commercial funding plays a larger role (e.g. pharma, medicine), but for most publicly founded research it's a rather bold assumption to claim that the funding is conditioned on achieving a particular result and not on the general impact of the results.