It is definitely a canard that all, most or even the majority of OA is Gold OA.
It is also definitely untrue that all, most or even the majority of Gold OA is APC-based.
But I think it is also true that the majority of non-APC-based Gold OA are not among the top journals in most fields — the ones most institutions need to subscribe to, and the ones that also tend to be the journals indexed by ISI (and that doesn’t just mean preoccupation with journal impact factors: those are also the journals that have established a track-record for high quality peer review standards).
I may be wrong, but think it is misleading to equate the canard about OA being Gold OA with the misimpression that most Gold OA is APC-based: It’s not, but there’s more to it than that.
And I also think that although it’s true that today’s limited and patchy Green OA has not caused journal cancelations, once OA becomes universally mandatory, Green OA will make subscriptions unsustainable, and journals will have to cut costs, downsize, and find another source of revenue to cover the remaining costs — and that other source of revenue will be Gold OA APCs, per paper submitted for peer review, at a fair, affordable, sustainable price, paid out of portion of each institution’s annual windfall savings from the subscription-cancellations induced by universal Green OA.
That will be affordable, sustainable Fair-Gold OA (as compared to today’s Fool’s Gold OA, double-paid alongside subscriptions at an absurdly inflated price). But I do not believe that either parallel subscription income, alongside universal Green, or subsidies, or (as some imagine) pure voluntarism and thin air will be sustainable ways of paying for the much-reduced but still non-zero cost, per paper submitted, of post-Green peer-reviewed journal publishing.