My personal experience only dates back one complete cycle prior to this one, so I have no information regarding the GRMs back in the early 1980s. Sorry.
Applying GRMs to single family residences is not normally considered a reliable or accurate method of valuation. Part of this is due to the variations of rationality involved at different times, and part is due to the difficulty in really pinning down the rental rates. The only reason I brought it up is because our original question involved trying to figure out how to establish a reasonable baseline.
Exploiting the relationships between income and sales price works quite well when the property types are typically rental-driven. Inasmuch as the commercial property types have generally followed the same trends as the residential, albeit at somewhat of a lag, the trends that are showing in the commercial markets can be compared to the residential trends as a secondary indicator. Commercial income multipliers have increased in the same way, and those increases are attributed to investor agressiveness and the expectations for continued price appreciation. Why else would an investor buy a property that doesn’t currently cash flow, if not for the expectation of making those losses up at the time of sale as a result of price increases?