It is a staple in the social ontology literature that social kinds have normatively loaded functions. In contrast to natural kinds such as water, money has a function to fulfil: It is supposed to enable widespread exchange of goods and services. Biological kinds might be used as an example of natural kinds with functions. However, functions of social kinds are of a stronger normative character that the evolutionary function of a heart to pump blood. While we might evaluate it negatively if a heart stops pumping blood, it is not bad from the evolutionary perspective, merely dysfunctional - not going to help to reproduce hearts. The idea that social kinds allow for normative evaluation can be found in Searle’s The Construction of Social Reality and has been broadly accepted.
A more recent formulation can be found in the paper “The Functions of Institutions: Etiology and Teleology” by Hindriks and Guala. They distinguish the etiological function, which is roughly equivalent to what I called “evolutionary function”, from the teleological function, which allows normative evaluation. According to them the latter relies on projection from the design stance by individuals.
What I have wonder is how we separate between the normative nature of social kinds and the normative force of social norms that concern social kinds but are not intrinsic to the kind. To see what I have in mind, it is best to first consider the case of the heart as a natural kind. It is quite reasonable to say that we have a social norm evaluating hearts not pumping blood as improper, worthy of our disapproval. That is not a feature intrinsic to the natural kind, because the social norms are not constitutive for the kind.
For social kinds, however, social norms might very well be constitutive. Consider the case of the kind US Dollar banknotes. One might argue that there exists a social norm that stores should accept these banknotes. A store in the US refusing to accept them might be worty of disapproval in virtue of this norm. But is this norm intrinsic to the kind? Is the fact the stores in the US should accept US Dollar banknotes intrinsic to the kind or not? I don’t know how to answer the question.
The problem is aggravated by the fact that social kinds change. Natural kinds such as water and hearts do not change, or at least I would assume as much. One might argue that the norm according to which hearts should pump blood would cease if the human population went extinct, even though the kind would still exist and be instantiated by organs of non-human animals. That line of argument does not apply to social kinds, at least not in a straightforward manner. One might argue that it was an intrinsic feature of US Dollar banknotes to be backed by Gold, but this is no longer the case. Not all intrinsic features of the kind might persist through transformations and this might also hold for normative ones.
If we cannot distinguish the normative aspects intrinsic to social kinds from those merely resulting from social norms, then the claim that social kinds have a normatively loaded functional nature becomes suspect. At least, it becomes hard to draw the line exactly. Are there intrinsic normative functions to the kind professor which prescribe certain action, or are these tagged on in addition as it were? Presumably one would have to assume a specific theory of social ontology to answer the question, but so far no one appears to have addressed it. For now, I will leave it at raising the question.
 I assume here an account of evolutionary functions along the lines of Ruth Millikan.
 Of course, that assertion has been questioned. Thomistic theories of natural law would be one example.
 “Intrinsic” appears more appropriate than “necessary” to me, but this linguistic intution might very well be questioned. I also hope to avoid discussing the modal character of kinds.
 I am not too committed at the relation being one of constitution. One might argue for grounding or any other ontological dependence relationship with explanatory force.
 At the moment there is no legal requirement for such acceptance, see: https://www.federalreserve.gov/faqs/currency_12772.htm
 Analysing this kind of change comes with challenges. One might, for example, wonder whether the kind exists in time and in fact undergo change or whether what changes is what kind we refer to when using the expression “US Dollar banknote”. On this second approach, kinds would remain unchangeable, instead language use would undergo transformation. I am willing to accept this second solution, but I speak loosely as if the first was the case.
 Please send me an email if I am wrong.