EDIT: to clarify, an argument that providing dad-jokes leads to less single motherhood (which then leads to less spending) is so tenuous, that if you accept it, you can accept almost any kind of mental gymnastics.
I agree that reducing single motherhood would probably lead to less spending. It's the connection to dad-jokes that tenuous (at least without any further evidence).
For most claimed benefits, it doesn't require a lot of mental gymnastics to find some arguments why they would reduce some spending in the long run.
If you give me an example as a challenge, I can (most probably) spin you the argument of how it reduces spending.
It's an argument against this specific argument for government spending.
Making bad arguments is excessively easy, independent of whether the thing argued for is good or bad, or true or false.
To give a simple example: consider a faulty proof of Pythagoras' theorem. See eg https://www.cut-the-knot.org/pythagoras/FalseProofs.shtml
Pointing out flaws in these purported proofs is not an argument against Pythagoras' theorem; it's still true after all. It's merely an argument against the faulty arguments.
I don't know whether it's a good use of tax money (probably not), but it's such a minuscule amount in the grand scheme of things, that it doesn't really matter.
Eg repealing the Jones act would do a lot more, if one wanted to get riled up about something. It's basically a very inefficient re-distribution from Hawaiians to a few domestic ship owners.