Quick attempt to summarise:

1. Earning to give is still impactful – probably more impactful than 98%+ of jobs. The current funding bar in e.g. global health by GiveWell is about 10x GiveDirectly, and so marginal donations still have about that level of impact. In longtermism, the equivalent bar is harder to quantify, but you can look at recent examples of what's been funded by the EA Infrastructure and Long Term Funds (the equivalent of GiveDirectly is something like green energy R&D or scaling up a big disease monitoring program). Small donors can probably achieve a similar or better level of effectiveness as GiveWell. Going forward, the crucial question is how quickly more opportunities at that level can be found. It might be possible to keep the bar at 10x, otherwise it's likely to drop a bit so that more funds can be deployed e.g. to ~5x GiveDirectly. If that happens, the value of earning to give in absolute terms will go down 2x, but still be very high.
    
2. Roles that help to deploy large amounts of funding above the current funding bar are more impactful than before (e.g. grantmaking, research, organisation building, entrepreneurship, movement building, plus supporting and feeder roles to these roles). This means their value has gone up relative to earning to give. (This is what we should expect because funding and labour are partially complementary, so as the amount of funding increases, the value of labour increases.) This means if you can find a good opportunity that's a good fit within one of these paths, it's seriously worth considering switching, and the case for switching is stronger than before.
   
   3. If you're earning to give and not going to switch, you could consider trying to add extra value by doing more active grantmaking (e.g. exploring new causes) rather than just topping up large pots. However, you still need to be able to do this better than e.g. the EA Funds, and it might still be more efficient just to earn extra money and delegate your grantmaking. Entering a donor lottery is another good option. It might also be better to focus on community building or gaining career capital that might make you happy to switch to direct work in the future (e.g. saving money).
   
    

It seems to me like the primary benefit of typical EA donors (say, most GWWC members, or anyone giving less than $100,000/year, i.e. the vast majority of us) giving effectively comes from the signaling effects of this behavior on helping to promote a culture of effective giving and effective altruism.

It still seems very worthwhile for typical EA donors like me to donate, since the direct value of my donations is still substantial and there's potentially this even greater signaling benefit on top of that.

That said, as Ben Todd summarizes in his answer, most EAs (i.e. everyone not in the reference class of people who have a nontrivial potential to become very wealthy EA donors) can probably do even more good through various kinds of work that help deploy the large amount of EA funding that already exists better and faster than they can through their modest donations.

Given that, I wouldn't want to encourage a small donor to donate a modest amount at the expense of them putting less time/effort/attention into shifting into a very valuable direct work career that helps deploy existing EA funds faster/better. But, if donating some percentage of a person's typical income helps keep them engaged with EA and thinking about important questions related to how we can all do the most good, then it definitely seems worth doing to me.

If anyone thinks I'm wrong about this, please let me know!

It it is not the full answer but in my experience active grant making is easier for a committed small donor than a fund.

At least two projects I am aware of: Effective Altruism London and the All-Party Parliamentary Group for Future Generations were made significantly more likely to happen by active grant making from small donors. In both cases the projects were being run by founders as volunteers. Donors who knew the founders and could form a view on their value of their work reached out and said: "this project is good, deserves full time staff and I could offer funding if needed". Both projects hired staff, grew and in later years went on to receive future funding from various official EA funds. I think in this way well-connected EtG’ers can make an huge impact that grant-makers cannot make.

I guess that, for me, donating is still morally better than things like buying wine. Plus, no headache.

Also, with Patient Philanthropy Fund, I guess it's unlikely that we can have too much funding.

I wrote a suggestion here, about donating to campaign contributions, which are capped in the US so many small donors is better than one large donor. https://forum.effectivealtruism.org/posts/FffuQRBYjvm5hiaFw/there-s-a-role-for-small-ea-donors-in-campaign-finance

I'm optimistic we will unlock new sources of needed funding (Rethink Priorities is working a ton on this) so we should expect the current funding overhang to be temporary, thus making it important to still have future donors ready / have large amounts of money saved up ready to deploy.

Additionally, I think many non-profits would benefit from increased donor diversity on top of the value of a marginal donation, since this helps more organizations with their stability by reducing idiosyncratic risk. And at least in the US it also helps with possible issues arising from the public support test.

Lastly, I think it would be valuable to have more active grantmaking / exploring / "giving to learn" from more donors. I got pretty involved in EA from doing this.