The earnings hit from recessions is short-term - this is the concept behind using smoothed valuation measures like the CAPE, though you can also just eyeball EPS trends and levels. In the GFC for instance, EPS had recovered to May 2008 levels (when the market was still only off around 10-15%) by Q3 2009. It did take about four years to regain the 2007 peak EPS (and a similar amount of time in the 1990 and 2000 recessions) but the fact remains that panicking out in late 2008 / early 2009 because earnings and dividends had fallen would have been short-sighted and foolish, an epic blunder. Obviously, the correct approach was to back up the truck and shovel every dollar you could into the market right as those earnings and dividends were plunging.
The same is going to be true in pretty much every recession / downturn, unless a) catastrophic policy mistakes or supply/demand shocks cause a Great Depression where GDP shrinks by 25%+ and doesn't recover, or b) the economy is so badly distorted that a return to normal conditions requires wiping out large swaths of the corporate sector, and abandoning activities which were artificially inflating GDP and corporate earnings to a huge degree - you saw this the PIIGS, especially Greece, Iceland, and Ireland where the economy became utterly reliant on huge sector imbalances, bubble activities and what amounted to accounting fraud, but it's much less likely to happen in an enormous highly diversified economy and stock market like that of the USA. All the more so if your portfolio has a fair bit of international exposure.
The other major exception is if valuations get so extreme (like the up to 100x PEs and negative carry on property seen in 1989 Japan) that there really isn't any meaningful equity risk premium, even compared to zero or slightly negative rates on cash. But we aren't there yet.
