It seems that the crucial thing here is which one can make a bigger impact. But since impact can be difficult to measure, some people focus on the exclusivity or difficulty in publishing as a proxy for quality, and quality is a proxy for impact.
And I suspect that books generally make a bigger impact for a range of reasons, not least of which is that “thought leaders” like to show off the books they’ve read and are influenced by. I’ve never seen, say, Bill Gates or Barack Obama hold up a journal article as something that has changed their thinking.
Imagine that you’ve written the best philosophy paper possible, and it’s published in the most “exclusive” journal in the discipline. If only 2-10 people read it, its impact would be highly restricted, and it becomes a real question of whether it was worth the effort. So, quality by itself might not be worth much, if there’s no audience, i.e., “Sleeping Beauty” papers.
Meanwhile, a book from a top publisher, e.g., OUP or MITP or trade presses like Penguin, that is of “lesser quality” (whatever that means) than the above journal article could still reach and move thousands and potentially 100,000+ readers, influencing many more minds. The best ideas and arguments don’t automatically win out in the marketplace for books, movies, and other content—just look at the idiotic TV shows, etc. that have been popular with the masses yet arguably have changed culture and the world, e.g., “The Apprentice.”
It can go the other way, too: a journal article can be very influential and cited widely, even if this is an impact that only academics notice. In that case, citation numbers would be a natural (but flawed) proxy for quality and impact. But this seems to be more the exception than the rule.
That’s a realpolitik answer. If the question is what’s more impressive to my tenure committee and dean who might not even be familiar with philosophy, then that may be a different conversation. But it still trades on the idiosyncrasies and prejudices of those people, just as the realpolitik answer does.
Anyway, “impressive” isn’t something that can be objectively measured, and impressing others is probably not a great thing to be motivated by, though I understand the concern for employment purposes. I’d say to just do you: if you don’t like doing things that are impressive to others but feel you must, that’s not going to be sustainable over time or likely result in your best work.