I have always taken the same position on that subject: the impact factor is not well understood. The article gets confused with the journal; that has been proven. It is connected with Lotka's law and is very simple to understand. I have made tonnes of speeches in the world. Often, scientists do not understand and review committees have to be told that they are prohibited from using impact factors, including the h index, which is false, since it is used differently by Google Scholar, Scopus and Web of Science.
So review committees have to look at CVs and not assume, as my colleague mentioned, that an article is necessarily good simply because it was published in Nature, for example. A lot of articles in Nature are never cited. It is not a synonym for saying it is good.
So the use of impact factors by review committees has to be prohibited. That is very easy to do. The three bodies simply have to adopt an internal regulation.