Have you ruled out one or more bad tubes and or something like a cracked parasitic suppressor or as you've said one or more out of tolerance screen resistors? The three final tubes are in parallel so if one goes South you'll only be able to draw 2/3 of the expected current which applies to bias current just as much as full output power current. Same thing if one of the parasitic suppressors is cracked at a solder joint or otherwise not supplying anode voltage. That one's a double whammy as you'll lose some of your current in all operating states but also fry the tube as the screen starts functioning as the anode. Screen resistors way out of value is a bit of a different beast but if they've drifted far enough it can impact your ability to draw expected current through one or more tubes. You can check for current balance or current at all through each of the three tubes independently by measuring the voltage across the cathode resistors one at a time. They may not be exactly equal due to variations between tubes, but assuming the resistors themselves measure equally the voltage drop across each when keyed in SSB mode should be close to equal with each tube carrying a third of the bias current. One of the cathode resistors is a bit tough to meter but not too bad if you turn the rig off, attach some clip leads to your voltmeter then turn the rig back on, warm it up and make the measurement.