The Keithley 2001 and 2002 are able to measure average ACV (though not converted to RMS value like most average-responding meters, so you'll have to convert). With the Keithley 2001-M firmware, the Keithley 2001 should also be able to measure RMS ACV based on the average AC voltage.
I also wonder if a good true RMS meter will not be able to reliable make this measurement, though again you will need to experimentally determine the conversion factor from results from an average-responding meter to the true RMS measurements.